It was more than 20 years ago. I had just gotten acquainted with Simpson's paradox. I was browsing in a bookstore and saw a book by an eminent statistician -- eminent in the sense that I had come across the name before through books and articles. The book was a collection of articles by him on many different topics in statistics. The very first article, or an article very early inte book after perhaps an Introduction, was on Simpson's paradox. The book was too expensive to buy. I had eagerly noted down the reference, but unfortunately I've lost it somewhere in all these years.
I hope one of you can pinpoint the title, based on these very vague tidbits about it:
I think the author was Irving J. Good, but now that I've done a considerable amoount of Internet search, I'm not 100% sure. The Internet and library search I did of titles by Good didn't turn up any such. But Good as author is somehow stuck in my mind. In any case, it was a statistician who wrote on myriad aspects of the field.
The entry had "Simpson's Paradox" in the title (that is, the title of the chapter; I, of course, don't remember anything specific about the title of the book!). In the 900+ pieces written by Good, there doesn't seem to be any. The closest was his his 1987 piece with Mittal: The Amalgamation and Geometry of Two-by-Two Contingency Tables.
The piece on Simpson's paradox started with an extensive discussion of whether it merits the label "paradox", as opposed to a riddle, an effect, a phenomenon, etc. I seem to remember that it concluded that designating it a "paradox" was justified, in view of the serious interpretational difficulties it caused.
It was a hardcover book!
Hints and speculations welcome.