William Sealy Gosset, the Guinness “Student” who invented Student’s t-test [0], was once thanked “for all [he had] done for the advancement of statistics”. His reply: "Oh, that's nothing - Fisher would have discovered it all anyway" [1].

Alas, Ronald Aylmer Fisher is long-dead and cannot discover anything anymore, so we’ll have to make do with whom is left - and that’s actually quite a lot. Many great minds have advanced statistics since the days of Fisher and Gosset, so the following three books can only offer a glimpse of this sprawling field of science with influences from economics to medicine, from sociology to astronomy, from computer science to biology, from mathematics to philosophy. But it is a glimpse worth taking, as the three author’s love and dedication for the field of statistics can be felt on every page.

For the Historically Minded: “The Seven Pillars of Statistical Wisdom” by Stephen M. Stigler [2]

Stigler has a skill that many of his colleagues lack: He writes succinctly and engagingly without oversimplifying things too much. By doing so, he gives both statistical laypeople and experts in the field something to work with. Laypeople can pick up “The Seven Pillars of Statistical Wisdom” to get an intuition about the basics of statistical thinking without being overwhelmed by formulae and scientific jargon. Experts will learn more about the long history of this type of thinking and the protagonists who have shaped it (Spoiler: It wasn’t R. A. Fisher alone).

For the Connoisseurs of Irony and Vitriol: “Dicing with Death” by Stephen Senn [3]

Senn knows how to hit where it hurts - and he doesn’t hold back. Be it headline-hungry journalists, rabble-rousing politicians or lazy lawyers: they all get their comeuppance - even the statisticians themselves. The harsh description of the abusive relationship between statistics and other sciences on the first page of “Dicing with Death” is a testament to this: “If mathematics is the handmaiden of science, statistics is its whore: all that scientists are looking for is a quick fix without the encumbrance of a meaningful relationship. Statisticians are second-class mathematicians, third-rate scientists and fourth-rate thinkers. They are the hyenas, jackals and vultures of the scientific ecology: picking over the bones and carcasses of the game that the big cats, the biologists, physicists and the chemists, have brought down.”

For the remainder of the book, Senn works hard at convincing the reader why everyone should have a more meaningful relationship with the “fourth-rate thinkers” - and he is not coy about it either: “If you think that statistics has nothing to say about what you do or how you could do it better, then you are either wrong or in need of a more meaningful job.”

Some chapters will be more difficult to understand for statistical laypeople, but you can enjoy this book even if you just skim over them. In any case, Senn’s cheeky account of what makes statistics such an intriguing scientific discipline is a treasure trove of apt quotes and entertaining insights.

For Statistical Footsoldiers and Veterans: “Statistical Inference as Severe Testing” by Deborah G. Mayo [4]

Mayo intends to “get beyond the statistics wars”. And she attempts to do so by providing a detailed account of the different reasons why those wars broke out in the first place. Whether she succeeds in brokering a lasting peace between the warring factions, I cannot tell. After all, I’m nothing more than a footsoldier in this conflict and as far as I can see, the statistical generals are still very much going at it on the scientific battlefield of the 21st century (aka “Twitter”).

What I can tell, however, is that “Statistical Inference as Severe Testing” has helped me tremendously with understanding - truly - some fundamental concepts of statistical inference in general and hypothesis testing in particular. Compared to the other two books mentioned, you need a deeper understanding of statistics to fully appreciate it, but if you are a statistically minded person who is anywhere close to a field in which hypothesis tests are routinely adopted, this book is definitely worth the effort.

References

[0] Servan L. Grüninger (2017). Wie eine Bieridee die Statistik revolutionierte. NZZ am Sonntag (https://www.servangrueninger.ch/blogcomplete/wie-eine-bieridee-die-statistik-revolutionierte, visited on October 20th 2020).

[1] Stella V. Cunliffe (1976). Address of the President, Miss Stella V. Cunliff, delivered to the Royal Statistical Society on Wednesday, November 12th, 1975. Journal of the Royal Statistical Society.

[2] Stephen M. Stigler (2016). The Seven Pillars of Statistical Wisdom. Harvard University Press.

[3] Stephen Senn (2003). Dicing with Death: Chance, Risk and Health. Cambridge University Press.

[4] Deborah G. Mayo (2018). Statistical Inference as Severe Testing: How to Get Beyond the Statistics War. Cambridge University Press