Review of Stigler (2016)

We just finished reading Stephen Stigler’s latest book The Seven Pillars of Statistical Wisdom (Harvard, 2016), pictured below. (Spoiler alert: Professor Stigler, a historian of statistics at the University of Chicago, summarizes the seven most revolutionary and counter-intuitive ideas in the history of statistics: such as the root-n rule, e.g. the idea that the second set of n observations is not as informative as the first set of n observations, and the concept of likelihood, i.e. the use of probability to measure the strength of an inference.) Here’s what we liked most about Stigler’s book (the good): his use of historical examples to explain the origins of each fundamental idea, such as an ancient Sumerian example of what today we would call “data visualization” (pp. 27-28) and “the trial of the Pyx” (pp. 47-50), a medieval test conducted by the London Mint to measure the accuracy of the weights of British coins. Here’s what we liked the least (the bad): Stephen Stigler is no Richard Dawkins or Nate Silver. That is, Stigler’s explanation of each statistical idea is hard to follow, and he makes no genuine effort to explain these important ideas to readers with no prior knowledge of statistics. Lastly, here’s the ugly: Stigler does not bother to cite the work on error statistics by Professor Deborah Mayo, even in his chapter on experimental design (Ch. 6).

Image result for seven pillars of statistical wisdom

Not suitable for Statistics 101.

About F. E. Guerra-Pujol

When I’m not blogging, I am a business law professor at the University of Central Florida.
This entry was posted in History, Probability. Bookmark the permalink.

1 Response to Review of Stigler (2016)

  1. Mayo says:

    Not being a Nate Silver when it comes to explaining statistical ideas can’t count as a criticism, at least as regards statistical inference. I’ve never, ever, anywhere seen someone write so fallaciously (yet so sure of himself!) vs Fisher and frequentism than Silver. He accuses Fisher of demanding immaculate data…among other pieces of silliness. I never could go back to much of the rest after that. It was harmful and deleterious, and I still await his correction on that. I will forgive him when he does. You don’t seriously mean that I should have been mentioned. Why? It’s enough that Stigler recognizes my hero, C.S.Peirce and spends time explaining how Peirce brought randomization into the field (back in the late 19th century), anticipating Fisher.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s