The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives (Economics, Cognition, And Society) - Book reviews of The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives (Economics, Cognition, And Society)

'The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives (Economics, Cognition, And Society)' book reviews

Review for'The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives (Economics, Cognition, And Society)'

"The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives" by Stephen T. Ziliak and Deirdre N. McCloskey is a meticulously researched and provocative critique of the widespread misuse of statistical significance testing in various fields, especially economics. The authors argue that an overreliance on statistical significance, particularly the p-value, has led to flawed decision-making with serious real-world consequences. Their thesis is both compelling and controversial, urging a rethinking of how data is analyzed and interpreted in scientific research and policy-making.

Ziliak and McCloskey begin by outlining the history of statistical significance and its incorporation into scientific research. They trace its origins to the work of Ronald Fisher and others, who developed the p-value as a tool for determining the likelihood that an observed effect is due to chance. However, the authors argue that this tool has been misapplied and over-interpreted in ways that often lead to misleading conclusions. They contend that the focus on whether a result is statistically significant (i.e., whether it passes a certain threshold, usually p < 0.05) distracts from more meaningful questions about the size, importance, and practical relevance of the effect.

The book is rich with examples and case studies illustrating the pitfalls of overemphasizing statistical significance. For instance, the authors discuss how this practice has led to misguided economic policies, such as the misallocation of resources in social programs or the erroneous interpretation of medical trials. They highlight cases where statistically significant results were touted as major findings, only for later research to reveal that the effect sizes were trivial or the results were not reproducible. This, they argue, has real consequences: wasted resources, misguided policies, and lost lives.

One of the book's strengths is its interdisciplinary approach. While it focuses heavily on economics, it also touches on applications in medicine, psychology, and other fields. This broad scope underscores the pervasive nature of the problem and the urgent need for reform. The authors call for a shift from a focus on statistical significance to an emphasis on economic and practical significance, advocating for a more nuanced interpretation of data that takes into account the context and magnitude of effects.

Despite its rigorous analysis, the book is not without its critics. Some may argue that the authors overstate the case against statistical significance testing or that their proposed solutions are impractical. However, even those who disagree with Ziliak and McCloskey's conclusions can benefit from engaging with their arguments. The book challenges readers to think critically about how they interpret statistical evidence and to consider the broader implications of their analytical choices.

In conclusion, "The Cult of Statistical Significance" is a thought-provoking and important contribution to the literature on statistical methodology and its applications. Ziliak and McCloskey's passionate critique of the current paradigm is a call to action for researchers, policymakers, and educators. By highlighting the limitations and dangers of an overreliance on statistical significance, the book encourages a more thoughtful and responsible approach to data analysis. Whether you are a seasoned statistician, an economist, or simply someone interested in the interplay between data and decision-making, this book is a must-read.

Similar BooksMore