As someone who last took a statistics course 35 years ago, I could never have foreseen the popular rage that statistics and data-crunching would become in current times (I still recall, and wrote about, my introduction to statistics as a youngster in a museum, though I didn't know it as such back then). And I haven't yet read Nate Silver's huge bestseller "The Signal and The Noise" -- I assume it is good and probably has a strong Bayesian tilt. But I love the volume I've just finished reading (which doesn't even mention Bayes): "Naked Statistics" from Charles Wheelan -- it's likely the best, most palatable introduction to statistics for layfolks I've ever seen. Not at all overly-technical, in fact sprinkled with fun and humor, and full of real-world examples (not abstractions) of statistical thinking in our day-to-day lives.
The book has a very nice progression of topic areas from means and medians and 'descriptive statistics,' through probability, correlation, the central limit theorem, hypothesis-testing, and on to regression analysis, but always with an emphasis on understanding underlying concepts, not specific empirical formulas or computation.
The author's focus is constantly on educating the reader as to why a basic understanding of statistics is vital self-defense in today's firehose world of information, journalism, science claims, and headlines. Clear examples are drawn from health reporting/diagnosis, gambling, Wall Street and the economic meltdown, polling, sports, business, and other everyday encounters (as well as including a very good chapter on the classic "Monty Hall Problem").
Some reviewers call Wheelan's approach "intuitive," which can be dangerous in so much as applied statistics and definitely probability can actually be very counter-intuitive at times, but again, this book is only dealing in the basics.
The one major drawback of the volume is that, as indicated above, there is no discussion of Bayesian analysis, which is sort of the 'golden boy' of much current statistical talk, that readers will miss out on.
My favorite chapter may be Chapter 6, describing how a probabilistic "Value at Risk" financial model from "overconfident math geeks" nearly brought down "the global financial system," but every single chapter is simultaneously intelligent and entertaining (...makes me want to read more of Wheelan).
The NY Times has a review from Abigail Zuger who rightly calls the book, "sparkling and intensely readable":
Zuger, an M.D., actually refers to it as, "the most important health book of the year... even though it’s not primarily about health," and then continues,
"...his multiple real world examples illustrating exactly why even the most reluctant mathophobe is well advised to achieve a personal understanding of the statistical underpinnings of life, whether that individual is watching football on the couch, picking a school for the children or jiggling anxiously in a hospital admitting office."
Wheelan incidentally, wrote a prior bestseller, entitled "Naked Economics," which I suspect is equally good if you care to brush up on that field of study.
As long as we're talking stats, some other recent pieces worth mentioning:
Another blogger believes, "All journalists should be required to pass a course in basic statistics before they are let loose on the unsuspecting public" and argues so here:
Meanwhile, Evelyn Lamb and Hilda Bastian are two bloggers who presented last week at the Science Online Conference in North Carolina (THE premier annual digital science communication gathering). Theirs was one of the few math sessions, and entitled: "Public Statistics: Blogging With Numbers":
Lamb wrote an introductory piece to the session for her Scientific American blog here:
...and Bastian covers statistics with a light touch at her blog here:
Finally, statistician William Briggs recently had Twitter attention tossed his way for an older provocative piece he did called, "Statistics is Not Math" in which he argues that "Statistics is not math; neither is probability... Statistics rightly belongs to epistemology, the philosophy of how we know what we know." (be sure and read the interesting comments as well):
Many people have recently made the case that statistics, in some form, should be part of the core math curriculum for ALL secondary students, and be a basic part of math literacy. Read the above books and links and you'll be on your way to having it covered. In fact, seriously, I think Wheelan's book (or something like it) ought be mandatory reading for all engaged citizenry!
In a 3-minute TEDTalk below Arthur Benjamin argues the case for required secondary statistics education:
And I'll close out with this example of probably how NOT to use statistics from another (tongue-in-cheek) 7-min. TEDTalk: