Statistics Done Wrong The Woefully Complete Guide

The book is widely praised for its clear, straightforward, and accessible narrative, making the often-complex world of statistical errors understandable and even entertaining. Reviewers highlight its relevance to scientific publishing and research, effectively illustrating the misuse of statistics through practical examples drawn from various fields, including medicine and social sciences. Many appreciate its witty tone and the systematic presentation of common pitfalls such as p-value misinterpretation, pseudoreplication, and publication bias. It offers valuable insights and practical tips, making it a useful resource for identifying and understanding flawed statistical practices.

Despite its strengths, a recurring point of criticism is that the book is not suitable for absolute beginners looking to learn statistics from the ground up. Many reviewers note that it assumes a foundational understanding of statistical concepts, and without this prerequisite knowledge, some explanations can feel terse or even confusing, particularly regarding complex ideas like p-values. Its relatively short length means it provides an overview rather than a comprehensive guide or in-depth solutions, leading some experienced statisticians to find much of the content already familiar. Additionally, a few readers found certain explanations insufficient or wished for more practical, hands-on advice.

Overall, the book is regarded as a valuable and thought-provoking read that successfully exposes the widespread issues in statistical practice within scientific research. It serves as an excellent primer for developing a more critical perspective on data analysis and published studies. The book is ideally suited for readers who already possess a basic grasp of statistics and are engaged in or regularly encounter scientific research. This includes working scientists, researchers, graduate students, and professionals in data-intensive fields like machine learning or bioinformatics who aim to identify and avoid common statistical errors and biases in their own work or in the literature.