8
$\begingroup$

Christmas is coming and I would like to make a statistics-themed gift. The recipient bought and liked How Not to Be Wrong by Jordan Ellenberg (btw, I like that book too). He also liked The Signal and the Noise, even though he found it a bit light on math. Thus I'm looking for something like these books, i.e. statistics-themed, not afraid of a few equations but lighter than a big caliber academic book such as for example BDA by Gelman et al. Can you suggest me a few titles?

EDIT: I just found out that he owns also Risk Savvy: How to Make Good Decisions by Gigerenzer but not Kahneman's book, so the suggestion by Glen_b seems spot on.

$\endgroup$
3
  • 2
    $\begingroup$ I look at the links to the two book. It doesn't help narrow thing very much. You could get 1 million answers. I like the one I am reading now. David Hand's "The Improbability Principle: Why Coincidences, Miracles and Rare Events Happen Every Day" Inexpensive paperback, thought provoking, alot of numbers but no equations. $\endgroup$ Dec 18 '16 at 20:13
  • $\begingroup$ @MichaelChernick, you say the links don't help narrow the search down much, but I think they did, since you proposed exactly the one book I thought of after writing the post...I don't what more to add to the description: pop-sci book about statistics/probability, not too simple, I gave an example of a book which actually shows some math, and of another book which instead shows very little math (even if it talks a lot abput forecasting). Some equations in the book would not be a drawback. $\endgroup$
    – DeltaIV
    Dec 19 '16 at 7:58
  • $\begingroup$ I think that given the spirit of the question, it's natural that multiple answers will be given. Anyway, if you have suggestions on how I could modify it to narrow down the range of admissible answers, I will implement them. $\endgroup$
    – DeltaIV
    Dec 19 '16 at 8:03
5
$\begingroup$

Dicing with Death by Stephen Senn focuses on medical statistics and is a lot more mathematical than The Signal and the Noise. I liked it but it does contain quite a lot of typos.

The Lady Tasting Tea covers a lot more ground than I expected and is one of the most open-minded statistics books that I have read. Although it is not at all mathematical, it does introduce a lot of interesting topics.

Symbols, Signals and Noise by Pierce is very cheap, easy to read and contains a lot of equations. It sounds like a great fit but is quite old (it's published as a Dover reprint.)

If your friend is interested in history, Games, Gods and Gambling by Florence Nightingale David is an account of the early history of statistics that is very heavy on the mathematics. I much prefer her writing to Stigler, but it's not really pop-sci, so maybe not suitable for your friend.

$\endgroup$
7
  • 1
    $\begingroup$ Salsburg;s The Lady Tasting Tea is a nice book. The title is based on Fisher's classic design problem to see if she can determine if milk is added before or after pouring the tea. $\endgroup$ Dec 19 '16 at 3:29
  • $\begingroup$ Very good! I would have gone with the first one, but if it's riddled with typos as you say, it may not be the right choice. I will check the other links. $\endgroup$
    – DeltaIV
    Dec 19 '16 at 8:24
  • 1
    $\begingroup$ Scrooge here. -1 on Salsburg. Full of wild errors. My review in Biometrics 57: 1273-1274 (2001) mentions several, but at the same time only some. (Reprints of Senn clean up a lot.) $\endgroup$
    – Nick Cox
    Dec 19 '16 at 17:35
  • $\begingroup$ @NickCox, so you're saying that, for what it concerns errors, "The Lady Tasting Tea" is worse off than "Dicing with Death", right? Did I interpret your comment correctly? $\endgroup$
    – DeltaIV
    Dec 20 '16 at 10:54
  • 1
    $\begingroup$ Yes. Senn understands history of statistics well. Salsberg knows lots of stories, but many are utterly wrong or confused. $\endgroup$
    – Nick Cox
    Dec 20 '16 at 10:58
4
$\begingroup$

On the lighter side of things? Statistics Done Wrong: The Woefully Complete Guide Is a fun title on, well, the title is self explanatory. In the same vain, a classic is How to Lie With Statistics. Similar-ish but with a slight Freakonomics vibe is Naked Statistics: Stripping the Dread from the Data

$\endgroup$
4
  • $\begingroup$ I checked and the first two were covered in Gelman's blog, the first one with good comments and the second one with negative ones. I'll look at the first and third one. Thanks!!! $\endgroup$
    – DeltaIV
    Dec 19 '16 at 8:17
  • 1
    $\begingroup$ Can you link to the negative comments on How to Lie With Statistics? I can only find some discussion of Huff's relationship with the tobacco industry, but nothing about the book per se. I've seen it highly recommended elsewhere - e.g. badscience.net/2008/01/the-huff $\endgroup$
    – nekomatic
    Dec 19 '16 at 10:20
  • $\begingroup$ An amazingly large fraction of cartoons in the original of Huff show smoking, including (you may not believe this, but it's true) babies smoking. Recent reprints in the US clean up some period language now deemed unacceptable, but leave the cartoons as they were. $\endgroup$
    – Nick Cox
    Dec 19 '16 at 17:38
  • 1
    $\begingroup$ @nekomatic uh! you're right. Gelman was talking about another book by Huff, How to Lie with Smoking Statistics, which never got published, but whose draft has been smashed by Alex Reinhart. The two books have a nearly identical name and the same author, so I was led in confusion. $\endgroup$
    – DeltaIV
    Dec 20 '16 at 12:38
4
$\begingroup$

Not at all mathematical, but certainly has some statistical elements:

Kahneman, D. (2011) Thinking, Fast and Slow

It's 5 years old so the person might just have read it but if they have not they might find it valuable.

$\endgroup$
1
  • $\begingroup$ This is also a nice idea! There are much more choices than I would have expected before asking my question :) $\endgroup$
    – DeltaIV
    Dec 20 '16 at 10:55
2
$\begingroup$

THE ACCIDENTAL STATISTICIAN an autobiography of G.E.P Box makes great (light) reading. It is not as heavy as the Ellenburg book which I just loved but enjoyable none the less as it describes the paths taken by all when trying to make sense out of numbers.

$\endgroup$
3
  • $\begingroup$ Are these good examples of the kind of book you want? $\endgroup$ Dec 19 '16 at 3:26
  • $\begingroup$ @MichaelChernick I think the best examples are "Dicing with Death" because it's relatively math-heavy (pity about the typos, maybe it has a site with errata and I could just print them out!) and the Pierce's book. Anyway, there were a lot of very good suggestions: I need to read more about "The Lady Tasting Tea", if it's thought-provoking it could be interesting even if not mathematical. $\endgroup$
    – DeltaIV
    Dec 19 '16 at 8:30
  • $\begingroup$ @MichaelChernick after reading your review on Amazon, I think "The Lady Tasting Tea" may be a good choice. I understand there are no equations, but still, a book which talks about "[..] p-values, Bayesian models, non-parametric methods, bootstrap, hypothesis tests and confidence intervals" sounds to me very mathematical, if not in form, for sure in spirit. $\endgroup$
    – DeltaIV
    Dec 19 '16 at 9:04
2
$\begingroup$

The History of Statistics: The Measurement of Uncertainty before 1900 by Stephen M. Stigler

I never read it back-to-back, just used it few times as a reference, but it's a nice book that could be interesting for a wide variety of readers.

$\endgroup$
1
$\begingroup$

For a physics flavor, Anomaly!, by Tommaso Dorigo is about the "discovery" of anomalous signals at several particle physics colliers in the 1990s, and the study of whether these were discoveries of new physics or statistical/methodological flukes. It is a bit pricey, however.

$\endgroup$
1
$\begingroup$

Sharon Bertsch McGrayne, The Theory That Would Not Die: How Bayes' Rule Cracked the Enigma Code, Hunted Down Russian Submarines, and Emerged Triumphant from Two Centuries of Controversy

Bayes' rule appears to be a straightforward, one-line theorem: by updating our initial beliefs with objective new information, we get a new and improved belief. To its adherents, it is an elegant statement about learning from experience. To its opponents, it is subjectivity run amok.

In the first-ever account of Bayes' rule for general readers, Sharon Bertsch McGrayne explores this controversial theorem and the human obsessions surrounding it. She traces its discovery by an amateur mathematician in the 1740s through its development into roughly its modern form by French scientist Pierre Simon Laplace. She reveals why respected statisticians rendered it professionally taboo for 150 years—at the same time that practitioners relied on it to solve crises involving great uncertainty and scanty information (Alan Turing's role in breaking Germany's Enigma code during World War II), and explains how the advent of off-the-shelf computer technology in the 1980s proved to be a game-changer. Today, Bayes' rule is used everywhere from DNA de-coding to Homeland Security.

Drawing on primary source material and interviews with statisticians and other scientists, The Theory That Would Not Die is the riveting account of how a seemingly simple theorem ignited one of the greatest controversies of all time.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.