Staff Pick: Standard Deviations

We are surrounded by data in many forms—patterns, signals, and information influence our decisions daily. Whether you work with data or not, understanding how it can be misused is essential. In Standard Deviations: Flawed Assumptions, Tortured Data, and Other Ways to Lie with Statistics, Gary Smith reveals how faulty statistical reasoning can distort our understanding rather than clarify it. I find his book illuminating, encouraging reflection on how we interpret and use data in our lives.

What makes Standard Deviations particularly illuminating is its discussion of two key ways we use data: to draw conclusions from patterns (inductive reasoning) and to test ideas or theories (deductive reasoning). For example, we often predict the future based on past trends, but if future conditions differ, these predictions can fail. Similarly, when testing a theory, it’s crucial to consider all the data, not just the parts that support our ideas. Smith highlights these common pitfalls, reminding us to approach data carefully and with an open mind.    

Working with data can be challenging because certainty is rarely guaranteed. Rather than accepting data at face value, we should treat data claims as testable hypotheses, not conclusions we’ve already made. Smith’s call for thoughtful analysis challenges us to adopt a more rigorous approach, addressing our biases, assumptions, and errors as they emerge. As data continues to play a central role in shaping decisions, Standard Deviations equips us with the tools to think critically and act responsibly. It’s time to stop passively accepting data and start questioning (in good faith, of course), analyzing, and using it with the scrutiny it deserves.