(Published in The Telomere Times)
“Do not kill people with your research.” So opens Charles Wheelan’s Naked Statistics: Stripping the Dread from the Data. Channeling the 10 Commandments through hyperbole-filled lines, Naked Statistics fills a critical knowledge gap in the scientific population through the much-needed translation of common statistics into layman terms. As up to one third of all published health science research papers have been noted as being unfounded and scientifically unjust, it is imperative that today’s researchers and researchers-in-training be well versed in all aspects of their discipline.
Following a logical progression seen in popular statistics textbooks, Wheelan successfully breaks down concepts such as the central limit theorem, correlation, and regression analysis in ways that scholars both young and old can pick up instantaneously. With an abundance of real-world examples, he illustrates the role of various statistical analyses in day-to-day life, as well as their inherent limitations. The chapter on common regression mistakes is especially effective at conveying popular misconceptions about the usage of regression analysis and the dangers of creating inferences where there are none. Journalists take note as here lies some of the best material in the analysis of primary sources.
The book does tend towards being overwhelmingly technical at times, pushing this popular science book into the realm of a textbook companion. That said, for astute readers looking to train their mathematical proof muscles, there remain plenty of resources in this book to do so.
If you are a fan of American sports, this book will resonate strongly with you as many of the examples come from basketball, football, and baseball. While not necessarily a wrongdoing on the author’s part, I believe this book would extend it’s reach to a wider audience if it incorporated larger global issues such as poverty, climate change, and market forces.
Overall, Naked Statistics does a fantastic job in introducing traditionally complex topics to a non-scientific audience. Better suited as a textbook companion than a subway read, I highly recommend this book to students currently engaged in entry-level statistics courses as the author’s ability to explain the relevance of the concepts they are learning far surpasses what any single teacher could hope to cover in the scope of a single lecture.