Rating: 
Summary: GOOD HISTORY OF STATISTICS AND ITS INFLUENCE IN SCIENCE
Review: This book is in no way technical or mathematical. The author focused on explaining basic concepts and their importance without getting into the details of the stats behind it. Hence, it is a book focused on a general audience wanting to learn about the history and the characters that pushed statistics forward mainly in the 20th century (there was not much before). It should be a very easy and interesting read for someone knowledgeable of very basic math and stats (if you know what a standard deviation is, you should be ok).
The author, as a lifelong statistician, is clearly in awe of the characters described, such as Pearson and Fisher, which do seem to a bit influence his writing. For example, he is descriptive of personal meetings with some of the main described characters, which leaves me thinking that, aside from a history of stats, this may also be his personal history. Overall, this is seldom seem, and it does not much hurt the content, which is clearly very well researched and written. If you are looking to add to your knowledge and are done with pop science, this may be an interesting next book.
Rating: 
Summary: This Book Has Been Overpraised
Review: This book is loaded with technical terms that are barely explained. True, the stories of the people involved in developing these ideas are presented in a vivid and clear style. But this does little to aid the reader in grasping these mathematical concepts so that they have meaning. After all, this book asserts that statistics has revolutionized science, and I did not feel anything like this at the end of this book. Moreover, the author glibly rejects all of chaos theory in a little bit more than 3 pages[93-6], at another point calls the law of cause and effect " a vague notion that will not withstand the batterings of pure reason"[ p 186] But what "pure reason" happens to be, the author never bothers to tell us. Moreover, he earlier asserts that "the statistical model that defines the quest of science...is also based on a statement of faith." This comment then merits a discussion of what axioms are being put forth. But again there is a conspicuous vacuum of thought.At the conclusion, Salsburg casts doubt on the entire subject of his book by stating :"I do not believe that the human mind is capable of organizing a structure of ideas that can come close to describing what is really out there." This self-contradiction coming from the mouth of a supposed science educator is a perfect reflection of the philosophical mess that constitutes this book. Not recommended.
Rating:
Summary: Stimulating, delightful, informative, mildly frustrating
Review: This is a delightful glimpse into the history of statistics that gives users and non-users alike a clear view of the perspectives and personalities that shaped the discipline. It will be enjoyable for most non-specialists and a must-read for anyone who has taken classes in statistics. The latter group will undoubtedly find the book somewhat frustrating, in that a few equations here and there would add significantly to Salsburg's descriptions of e.g. the elegance of certain distributions or the spare beauty of others. Avoiding math was undoubtedly the right decision from a mass-market point of view, but (enterprising authors, take note!) a companion volume with all of the equations would make this one immensely appealing for introductory statistics classes.My only real complaint is that the narrative style is at times rather abrupt. It could hardly be otherwise, given the task that Salsburg sets for himself. Imagine, for example, a book on the history of cooking that described the evolution of new dishes without including any recipes: the author would be limited in how much he can describe, and when he moved on inquisitive readers would be left thinking, "Well, wait, now -- how much basil? How much garlic?" etc. Similarly, the limits on the amount of detail in which the author can describe advances in statistics often left me feeling as though we'd moved on when only part of the story had been told. I can't imagine how it could be otherwise, though, and to give credit where credit is due the author has a genuine knack for boiling complicated statistical issues (e.g. controversies surrounding the Neyman-Pearson approach to hypothesis testing) down into very straightforward language. The final chapter is stimulating in a big-think sort of way: Salsburg, having documented the development of statistics over the past century, reviews its sometimes questionable foundations and suggests that before too long it may be replaced with something else. What that "something else" might be is tantalizingly vague, but the chapter serves as a useful reminder that revolutionary modes of thinking, like those that started the statistical revolution, are always waiting in the wings.
Rating: 
Summary: Good for non-statisticians
Review: This is a pretty good book for non-statisticians, but then why would a non-statistician ever want to read it? As an applied statistician I was repeatedly frustrated by the high-level summaries of most topics with no real depth. For example, the section that discusses bootstrapping was so cursory that without prior knowledge and experience I would not have had a clue what bootstrapping is or how it works. I know the target audience was the general public, but in some cases a little more depth would have moved this from being only a good to a great book. One thing that I found amusing was the cover art--didn't the graphic artist even read the book? If you read the story that is the basis for the book title you would NEVER have shown a cup of tea with lemon. One does not use lemon and milk in tea. If you are looking for an excellent book that covers the role statistics has played in industry and science and has some depth, consider George Box's "Box on Quality and Discovery" ... This book is approachable for non-statisticians and will keep us statisticians awake!
Rating: 
Summary: The Lady Tasting Tea -- Delicious!
Review: This is a readable, enjoyable yet thought-provoking book that shows the impact of statistics and statistical thought on science. It does so by using a variety of examples including many from daily life that any reader can relate to. (What does a 95% chance of rain REALLY mean anyway?) By giving a vivid picture of individual statisticians as people, the book also shows the ways in which individual personalities play a role in the development of ideas and in the mentoring of new generations. This book was definitely a 5 star addition to my library.
Rating:
Summary: Interesting history of statistics over the 20th Century
Review: This is a really nice history of statistics over the 20th Century and the impact of statistics on advances in science and medicine. I particularly enjoyed many of the interesting anecdotes of the people involved in the development of statistics. My only complaint is that this book is very good in describing the input of East Coast statisticians but is rather lacking with respect to West Coast statisticians. West Coast statisticians are either not mentioned, or their affiliations are incorrectly stated. I was somewhat surprised that others had not noted this glaring gap, which is why I wrote this particular review and give the book only 4 stars. Otherwise, this is an interesting, entertaining read.
Rating:
Summary: The Lady is Great!
Review: This is a terrific book. Salsburg makes the history and some of the key principles of statistics come alive using simple, non-mathematical language. This is the only book I have ever read that could seduce the innumerate into the love of numbers and mathematical reasoning.
Rating:
Summary: Where's the beef?
Review: This is another of those books about mathematics that avoids displaying any mathematical equations or formulas, like Stephen Hawking's "A Brief History of Time." Hawking explained that his publishers told him that every line of mathematics would cut the sales by 50%. Unfortunately, the lack of mathematics makes it much harder to transmit the ideas. In many places in this book, I miss the math sorely. This book is actually harder to understand as it is, then if it gave us the relevant equations and formulas. Maybe Salsburg (or some other generous soul) could put together a web site that provides the math, referenced to chapters in the book. Nevertheless, the book tells its stories well, and rewards (and challenges) the reader with some ideas and concepts I'd like to at least remember, if not explore further. Chapter seven briefly outlines "the Fisherian versus the Pearsonian view of statistics," a philosophical theme that is fundamental to the application of statistics to reality. "Karl Pearson viewed statistical distributions as describing the actual collections of data he would analyze. According to Fisher, the true distribution is an abstract mathematical formula, and the data collected can be used only to estimate the parameters of the true distribution.... Pearson viewed the distribution of measurements as a real thing.... To Fisher, the measurements were a random selection from the set of all possible measurements." I'm not sure what this means, exactly. At the end of chapter six, there is a review of a math text (Gumbel's "Statistics of Extremes," an out of print collectible being offered through amazon...) from the 1950's that makes me want to find the book and read it. The appreciation for this text is put into context by a quick sketch of the varying strengths and weaknesses of the "definitive" books in the entire realm of mathematics. Chapter ten begins with the surprising statement, "Chaos theory is actually an attempt to undo the statistical revolution by reviving determinism at a more sophisticated level." Salsburg actually brings in the subject of chaos theory only as an introduction to the statistical concept of "goodness of fit." World War II stories abound in this book, as it had significant impact on the lives of most of the people who brought statistics forward in the twentieth century. Nazism and fascism made martyrs of some statisticians, interrupted the careers of some, accelerated the careers of others, and drove quite a few to the United States. Chapter eighteen provides an overview of how the link between cigarette smoking and lung cancer came to be accepted, and how the question itself drove the discipline of statistics. It also highlights a fundamental question first exposed by Bertrand Russell regarding cause and effect. In chapter twenty-four, we learn how Japanese industry became known for superior products (notably automobiles) by heeding the advice of W. Edwards Deming, an American statistician who was unable to attract the right kind of attention in the U.S. The final chapter steps back to get the big picture of philosophical questions that may signal the limitations of statistics as we know it, limitations that may someday be the seed of a future intellectual revolution that will bring forth an entirely new science that transcends statistics altogether. (1) Can statistical models be used to make decisions? (2) What is the meaning of probability when applied to real life? (3) Do people really understand probability? It may not be immediately apparent that there's anything worth serious consideration in these questions, but Salsburg brings them to relevant life in just a few pagess. Brief, insightful observations on the nature of mathematical activity and mathematical community are scattered throughout the book, as if they were salt seasoning a dish to make it tastier. He notes that mathematical research rarely bears fruit without the cross pollination of discussion among mathematicians. These discussions help to expose both errors and new ideas. He has fun with the semi-facetious "law of misnomy," which claims that mathematical ideas and techniques are generally not named for the person who actually created them, but for someone else. I enjoyed this book and learned a bit from it, but the absence of mathematics is a major shortcoming. Who would want to read a history of statistics if they weren't mathematically inclined enough to benefit from a few lines of greek letters and subscripts?
Rating:
Summary: Math is never easy.
Review: Writing a story of mathematics for a general audience is never easy (in addition to this book I am also thinking of Berlinski's "Tour of the Calculus"). Salsburg aims to provide a broad, non-technical overview of the development of modern statistical theory. The story moves from the early 20th century, the days of Pearson and "Student's t" to Deming and modern computer-aided analysis. By his own admission there are gaps in the narrative -- he focuses on those areas where he, as an academic, already has some familiarity. He does touch on the core elements of basic statistics -- p-values, t-test, hypothesis testing -- but his work, despite the complete absence of mathematical notation, still requires more than a casual knowledge of the topic. For example, he takes for granted that the reader knows what a "normal distribution" is. That being said, a person with at least a basic knowledge of statistics (maybe even one semester in college) is the most likely audience for this work, and would likely understand some of the basic concepts Salsburg neglects to explain.
| 
