Statistically speaking

Particle physicists use statistics to distinguish new particles from meaningless blips. Climate physicists need it to turn incomplete data into robust models. And pretty much everyone in science wheels it out to calculate error bars. Despite its usefulness, however, the field of statistics has a reputation for being dull and confusing; as a former classmate of your reviewer’s put it, “If I had one day left to live, then I would live it in stats class, because that way, it would last longer.” In his book Naked Statistics: Stripping the Dread From the Data, author Charles Wheelan concedes that his subject could do with a bit of a boost. The statistical sexing-up begins with the book’s front cover (which depicts a cartoon character holding a suggestive and strategically placed bar graph) and continues with the kind of writing that made Wheelan’s previous book, Naked Economics, into a bestseller despite its “dismal science” subject. In addition to clarity, Wheelan’s favourite tactics for keeping the reader hooked include a nice line in silly examples and a ready spoonful of wit to make the statistical medicine go down. In the second chapter, for example, he tells a story about a bar patronized by Bill Gates, a talking parrot and 10 low-paid workers. The point of this anecdote is that mean values can be distorted by outliers; when Gates walks into the bar with the parrot on his shoulder, the average annual income of the patrons leaps from $35,000 to about $91m. (The parrot, Wheelan writes, “has nothing to do with the example, but it kind of spices things up”.) He keeps the tone light even as the book moves on to more complex topics, such as the “garbage in, garbage out” problem of doing statistics with bad data and the various biases that can lead people to form erroneous conclusions. At the heart of the book is the central limit theorem, a pillar of statistics that, crudely speaking, states that a random sample of a given population will resemble the population as a whole. That might not sound terribly earth-shattering, but Wheelan shows that if properly applied, the central limit theorem can do some amazingly useful things – such as detecting when someone’s been fiddling experimental data. The theorem also allows us to infer that while most physicists won’t learn much from the first few chapters of Naked Statistics, nearly everyone will get something out of it by the end – a pretty good result, whichever way you slice it.

• 2013 W W Norton/Wiley £18.99/$26.95hb 320pp

A classy bunch of particles

Following the success of Brian Greene’s The Elegant Universe, the word “elegant” has been somewhat over-used in popular-science writing. However, there is no other word to describe Jeremy Bernstein‘s A Palette of Particles: this is an elegant book, elegant in its writing and in its packaging, the sort of book that ought to be read with freshly washed hands and a glass of something cold and sophisticated. Its subject, broadly speaking, is the history of particle physics, and it is organized into three sections. The first of these deals with the electron, proton, neutron and neutrino, which Bernstein describes as the “primary colours” on nature’s palette. Next up are more exotic particles such as the pion and muon, antiparticles, strange particles and quarks; these are the “secondary colours”. The final section is devoted to nature’s “pastels”, beginning with the relatively vibrant Higgs boson and shading away to shadowy objects such as tachyons and gravitons. Bernstein is both a physicist himself and a veteran observer of other physicists; in many cases, he met or was personally acquainted with the people he describes, including Wolfgang Pauli, Emilio Segré and Sheldon Glashow. Perhaps as a result of this familiarity, the book sometimes reads as though it was written not for its stated audience (“a general reader with an interest in science”), but for insiders who, like the author, are already familiar with the main characters and their histories. It is, if you like, a bedtime story for quantum connoisseurs rather than a bracing wake-up call for newcomers – and like the best bedtime stories, it offers plenty for readers to think about. In Bernstein’s view, “nature [is] something like those Russian dolls, except in this case I think there may be no limit. The deeper we probe, the more will be left to probe.”

• 2013 Harvard University Press £14.95/$18.95hb 224pp