Alan B. Krueger (1960–2019) was an influential American labor economist at Princeton University who also served in high-level policy roles, including as Chair of the Council of Economic Advisers under President Obama. His work is central to Naked Statistics as it provides key case studies that demonstrate how careful statistical inference can overturn conventional wisdom. Wheelan, who was Krueger’s former student, uses his mentor’s research to bridge the gap between academic theory and policy practice, positioning Krueger as a credible authority on applied econometrics.

Krueger’s work is used to illustrate the critical difference between correlation and causation. His book, What Makes a Terrorist (2008), which found that terrorists are more likely to be educated and from middle-class backgrounds, serves as a powerful example. Wheelan presents this finding to show that a simple correlation between poverty and terrorism is misleading and that careful statistical controls are necessary to uncover the true drivers of complex social phenomena. Krueger’s work exemplifies the kind of statistical detective work that challenges assumptions and informs policy.

Kruger’s research with Stacy Dale on the long-term earnings of college graduates is another cornerstone of Wheelan’s argument about program evaluation. By tracking the outcomes of students who were accepted to elite colleges but chose to attend different institutions, Krueger and Dale created a study that addresses the problem of selection bias. Their finding—that the earnings premium from attending an elite school only exists for students from poor backgrounds who were accepted but went elsewhere—demonstrates how a well-designed study can approximate a counterfactual scenario and isolate a specific causal effect. This work showcases a powerful method for making credible claims without a fully randomized experiment.

Esther Duflo (1972–), an MIT professor and co-founder of the Abdul Latif Jameel Poverty Action Lab (J-PAL), is a leading figure in development economics and a central exemplar in Naked Statistics for the power of program evaluation. A co-recipient of the 2019 Nobel Prize in Economic Sciences, she championed the use of randomized controlled trials (RCTs) to test the effectiveness of anti-poverty interventions. For Wheelan, Duflo’s work embodies the gold standard of causal inference, demonstrating how field experiments can provide clear, actionable evidence for policymakers.

Wheelan highlights Duflo’s methodology through concrete examples. Her studies, such as providing camera-verified incentives to reduce teacher absenteeism in India or offering small nudges to encourage fertilizer adoption among Kenyan farmers, illustrate how randomization can cleanly isolate a treatment effect. By creating distinct treatment and control groups in real-world settings, her experiments avoid the selection biases and confounding variables that plague many observational studies. These RCTs serve as the clearest examples in the book of how to establish a causal link between an intervention and an outcome.

Ultimately, Duflo’s contribution to Wheelan’s argument is both methodological and philosophical. Her approach of breaking down large, complex problems like global poverty into smaller, testable questions supports the book’s thesis that rigorous, evidence-based analysis leads to better solutions. Her Nobel Prize validates this experimental approach, reinforcing Wheelan’s claim that well-designed statistical studies are among the most powerful tools for improving human welfare.

Sir Michael Marmot (1945–) is a British epidemiologist at University College London whose long-running Whitehall studies of British civil servants provide a foundational case study for the regression chapters in Naked Statistics. Beginning in the 1960s, Marmot’s research linked social factors, particularly job control and status, to health outcomes like coronary heart disease. His work was pivotal in establishing the field of social determinants of health and serves as Wheelan’s primary example of how multivariate regression analysis can uncover complex relationships in data.

The Whitehall studies exemplify the power of longitudinal cohort analysis. By following thousands of civil servants over several decades, Marmot was able to use regression to isolate the impact of psychosocial work factors on health. Wheelan uses this research to demonstrate how statisticians can control for confounding variables. Marmot’s analysis showed that even after accounting for traditional risk factors like smoking, diet, and blood pressure, low job control remained a significant predictor of heart disease. This illustrates a core function of regression: separating the influence of one variable from others to avoid spurious correlations.

Marmot’s work supports Wheelan’s broader argument about the importance of careful statistical design in making credible causal claims from observational data. The finding that social hierarchy itself has physiological consequences was a landmark discovery for public health policy. For Naked Statistics, it serves as a powerful, real-world demonstration of how a well-specified regression model can reveal subtle but profound truths about society and human health.

Nassim Nicholas Taleb (1960–), a scholar and former derivatives trader, is presented in Naked Statistics as a crucial critic of conventional financial risk models. He is best known for popularizing the concept of “black swan” events—highly improbable, high-impact occurrences that standard statistical models fail to predict. Wheelan uses Taleb’s critique to anchor the book’s cautionary chapter on the misuse of probability. Taleb argues that many financial models, such as Value-at-Risk (VaR), create a false sense of security by underestimating the likelihood and impact of extreme events. His focus on “fat-tailed” distributions, where rare events are more common than a normal distribution would suggest, explains why models based on past market behavior failed so catastrophically during the 2008 financial crisis. Taleb’s work serves as a powerful argument for intellectual humility, supporting Wheelan’s central theme that statistical precision should never be confused with accuracy.

Adriana Lleras-Muney, a UCLA economist, provides a key example in Naked Statistics of a natural experiment, a powerful tool for establishing causality without a randomized trial. Her research used historical changes in state-level compulsory schooling laws as an instrumental variable to isolate the causal effect of education on adult mortality. Because these law changes forced some individuals to get more schooling than they otherwise would have, they created a quasi-random variation in educational attainment. Lleras-Muney found that an additional year of schooling significantly reduced adult mortality rates. For Wheelan, this study is important because it broadens the reader’s toolkit for causal inference. It demonstrates how clever research design can leverage policy changes or other occurrences to overcome the limitations of observational data, offering a credible alternative when RCTs are not feasible.