Though I am someone who appreciates numbers and statistics, I decided to read this book to ensure I am not trapped by the dogma that everything carrying a numerical tag is inherently true.
In the first chapter, the author traces the origins of modern statistics and data representation back to the Crimean War of the 19th century when Florence Nightingale compiled data on the number of wounded and dead soldiers, persuading authorities to allocate funds and improve amenities. It is true that numbers can save lives, but it is also important to recognize that they can often be misleading and lead to a misinformed narrative. Sanne Blauw highlights the IQ study among different ethnic groups to support this notion, emphasizing that value judgments vary among populations, and relying solely on standardized IQ tests can limit the outcomes. Moreover, using a single metric to assess a complex parameter like intelligence has inherent limitations.
In another chapter titled "The Shady Sex Study," the author explores the importance of sampling, as a smaller sample size cannot truly represent the entire population. This issue is also evident when data from Western, Educated, Industrialized, Rich, and Democratic (WEIRD) countries is generalized in the field of psychological studies. Additionally, she points out that survey questionnaires can often be leading in nature, causing respondents to unknowingly align with the interviewer's narrative.
Establishing causality from mere correlation is a complex matter, particularly in the field of medical sciences. The author provides a historical note on how cigarette manufacturing companies evaded the truth about tobacco's carcinogenic potential. In this context, understanding the difference between relative risk and absolute risk is crucial. For instance, some vegans argue that consuming meat increases the risk of cardiovascular diseases without considering that relative risk (e.g., 20%) differs from absolute risk (e.g., 1 in 100). Similarly, the correlation does not imply causation when individuals claim that consuming wine is beneficial for health. While resveratrol in wine is bioactive, the Mediterranean diet is a significant factor in improved heart health among populations in Italy, Spain, and other regions.
Towards the end of the book, the author discusses three important fields where numbers play a dominant role and will shape the future:
1. Credit systems
2. Predictive policing
3. Health insurance
While we may feel enthusiastic and pleased that algorithms and predictive modeling will replace human subjectivity, it is crucial to question how we will utilize these algorithms. Upon finishing the book, I came to realize that it is our psychology that assigns value to numbers, and it is essential to understand and overcome our internal biases when analyzing numerical data.