- Statistics Without Tears – A Primer for Non-Mathematicians
- by Derek Rowntree
- Charles Scribner’s Sons
- Copyright 1981
- 190 pages
Rating (for the entire book) – 7.5 out of 10
Well, I finished Statistics Without Tears Wednesday morning. I’m glad that I read it, but I’m ready to move on to something new. Here are my thoughts on the end of the book –
Chapter 7 (pp. 127-154)
Fortunately, Chapter 7 is better than is Chapter 6. Consistent with the rest of Tears, Rowntree covers a TON of ground in just a few pages. A thought that occurred to me while reading Chapter 7 was that the book would have been better if I’d taken it in smaller doses, so that I could have pondered what Rowntree was discussing.
Among many other topics, Chapter 7 includes a fantastic discussion of one- and two-tailed significance tests. Rowntree also covers t-tests, F-tests (ANOVAs), and Chi-Square tests. I followed pretty well up to the Chi-Square tests, which seemed to be one topic too many. Luckily, I had a decent understanding of Chi-Square tests before I started the book.
But my favorite part of the chapter was the end where Rowntree promises “You will no doubt have found the last two chapters to be tough going. The next, and final, chapter should seem considerably easier” (p. 154). Sweeter words I’ve never read.
Chapter 8 (pp. 155-184)
Rowntree is good to his word, but the pace never slackens. Chapter Eight covers both correlation and regression in less than 30 pages. The correlation material is excellent, particularly his explanation of reliable and unreliable correlations (p. 167). The last five pages of the book finally get to prediction (i.e., regression). Rowntree does will with his “quick-and-dirty” description.
Postscript (pp. 185-190)
The Postscript begins with a nice flourish, as Rowntree states that his goal in Tears is to provide “…a bird’s eye view of the field of statistical concepts, rather than [using traditional methods, which] take you crawling, like a calculating snake, through the undergrowth of statistitical computation” (p. 185). To me, “a bird’s eye view” is a perfect description of this book.
However, after this cool passage, Rowntree closes with a whimper. He provides a too-brief summary of too many concepts. Then he closes with some warnings on the ways that statistics can mislead us. The warnings are well taken, but such a well-written book needs to close with a flourish.
Eating My Veggies
When I do my professional reading, it always feels like eating my veggies – good for me, but not necessarily what I want to do. I mentioned this to my wife earlier today and she said, “Yes, but some authors sautée the veggies in butter and add some cheese.”
Indeed, that’s a good description of Statistics Without Tears – you’re still eating veggies, but they are prepared in the tastiest way possible. I recommend the book for those who want to better understand basic statistics.