PDFLINK |
New and Noteworthy Titles on our Bookshelf
February 2024

Probably Overthinking It
How to Use Data to Answer Questions, Avoid Statistical Traps, and Make Better Decisions
By Allen B. Downey. University of Chicago Press, 2023, 256 pp.
Courtesy of Chicago University Press.
In the age of Big Data, we might take for granted that we can find the answers to our questions within a data set. While there is some truth to this, we must also acknowledge our limitations in understanding the data. There are paradoxes that we could succumb to without proper investigation. Probably Overthinking It started as a blog and is now compiled into a book that discusses many data sets and possible ways to analyze them. Downey explores many data sets: family size, recidivism rates, COVID-19 superspreaders, smoking and birth weights, and ratings of chess players, just to name a few. He discusses distributions and possible reasons why data sets form a certain type of distribution.
Perhaps the most interesting component of the book is the discussion of paradoxes. I learned about a selection bias called length-biased sampling which can result in paradoxical conclusions, such as concluding that there are no cars driving your speed on the interstate, only cars driving much faster or much slower than you. Downey provides many examples of Berkson’s paradox and Simpson’s paradox. If you cover these in a data science or statistics course, you might be interested in this book. You could also analyze the data sets yourself or have students explore them; Downey invites anyone to use the code he wrote for this book and points readers to relevant and available data sets. The book contains helpful data visualizations without too much technical detail, which should allow mathematicians in many fields to benefit from this book.

“You Are Not Expected to Understand This”
How 26 Lines of Code Changed the World
Edited by Torie Bosch. Princeton University Press, 2022, 216 pp.
Courtesy of Princeton University Press.
Our lives are so integrated with technology that it is hard to imagine life without email or the internet. We long for “likes” and put up with pop-up ads. You Are Not Expected to Understand This contains 26 essays about pieces of code that have changed the world in unexpected and extensive ways. The essays are historical as well as anecdotal, and they span from the origins of coding to modern viral images. Many of the essays are authored by those who were on the front lines of coding changes, and their contributions to the book are short, only six or seven pages long.
I was amused when I learned that the popularity of the Roomba (the robot vacuum) is attributed to its slightly chaotic path around a room. I also gained cultural insights when reading essays about the logistical challenges of the gender spectrum when many databases require a binary entry and how coders put Bangla, the language of Bangladesh, online to help more of the world communicate. I appreciated that the book highlights flaws in code. Coders are human, and rather than expecting them to be infallible, we should understand the limitations of their code. I expect that many mathematicians who lived through some of the events discussed, such as the creation of hyperlinks or the JPEG, would appreciate this book. I also think a younger generation that has grown up with the consequences of the choices of programmers would gain insight as well.