Last week, I came across three intriguing statistical observations: "[A major survey] shows the pilots' mortality rate was 61 per cent of that for men in England and Wales overall"; "17 per cent of patients surveyed in a hospital had been waiting for more than four hours"; and "children who are kidnapped are far more likely to be whisked away by relatives than strangers, just as most people are murdered by people they know".

When you think about them, none of these makes sense. Since the overall death rate is undoubtedly 100 per cent, the first statistic claims that 39 per cent of pilots live for ever. Presumably some vital component is missing, such as the phrase "under the age of 60". From the figure quoted for hospital queues, we might infer that a patient has a 1 in 6 chance of being made to wait for four hours, but if this is the result of a snapshot of all patients waiting at a given moment, then it will greatly exaggerate the numbers kept waiting. Anyone kept hanging around for four hours is more likely to be included in such a survey than a patient seen promptly. Without more information on the methodology, the figure tells us nothing. Lastly, that little item about most people being murdered by people they know. Most people aren't murdered at all. Most people die from natural causes.

Such examples show how much most people would benefit from reading The Universe and the Teacup. Subtitled "The Mathematics of Truth and Beauty", this book is a romp through the world of mathematics and science, leaving scarcely any areas unromped. The opening chapters, an account of scientific method and a plea for numeracy, serve as a springboard for a dive into astronomy, particle physics, quantum mechanics, biology, psychology, linguistics and mathematical sociology, and a final section that touches on the structure of the universe and the nature of scientific truth.

The scope is formidable, yet KC Cole flits through it all with a remarkably light touch. There is only one equation in the entire book, Einstein's "e = mc2", which a footnote tells us means "energy equals mass times the speed of light (c) squared". Yet if the author feels she has to expand, in such unhelpful terms, what such an equation means, then there can be no hope of explaining the real complexity and excitement of scientific discovery.

This is a science in soft focus and occasionally out of focus - as in one of the examples quoted at the beginning of this review. The line about most people being murdered comes from this book. Scientific method is, above all, about rigour, and science writers, more than anyone else, ought to avoid sloppy language. Despite all its defects, however, most readers will find this an entertaining and enlightening book. It is good to know that you cannot have a teacup the size of a planet (because the gravitational pull of the cup would make the handle collapse), and to learn that the sentences "time flies like an arrow" and "fruit flies like a banana" pose problems for linguists; but anecdotal science is no substitute for the real thing.