A fire hose of information bombards us every day, and some of it is even true. Is Bitcoin a good investment? Are hurricanes getting worse? Is the measles vaccine dangerous? Separating the wheat from the chaff is what statisticians do鈥攁nd there鈥檚 lots of chaff. To Think Like a Statistician shares the skills statisticians use to sift through evidence, learn from experience, and extract meaning and knowledge from the random and the contradictory.
Bradley Efron is one of the most renowned statisticians in the world and has shaped how data science and machine learning are practiced today. In this book, he draws on examples ranging from David Hume鈥檚 critique of miracles to counterfeit Basquiats, AI hallucinations, pandemics, competing political claims, government approvals of Alzheimer鈥檚 treatments, gambling, and misinformation. He describes how statisticians have tackled difficult topics鈥攍ike correlation, causation, prediction, survival, and accuracy鈥攁nd demystifies the disputes surrounding concepts like randomness, uncertainty, and subjectivity.
Blending real-world insights with personal stories from a leading expert, To Think Like a Statistician equips readers with powerful ideas from the statistician鈥檚 toolbox and explains the tricks of the trade, enabling anyone to become a more sophisticated consumer of information in an increasingly noisy world.
Bradley Efron is the Max H. Stein Professor Emeritus of Statistics and Biomedical Data Science at Stanford University and a recipient of the National Medal of Science. His books include Exponential Families in Theory and Practice and Large-Scale Inference. He is a MacArthur Fellow and member of the National Academy of Sciences.
“In this delightful and surprisingly funny read, Efron shows us—rather than tells us—the role that statistics plays in the world around us. If this were required reading for eighteen-year-olds, they’d all become statistics majors!”—Daniela M. Witten, University of Washington
“An entertaining and thought-provoking book by one of the most creative and important statisticians of the modern era. I recommend it not despite but because I disagree with it in so many places. It is from tough problems and controversies that we learn the most.”—Andrew Gelman, Columbia University
“An engaging and accessible exposition of statistical ideas by a leading creator of modern statistical science, showing how and why statistics is central to our understanding of the world.”—David Hand, Imperial College London
“There is no shortage of popular books on quantum mechanics, string theory, black holes, and the like. But popular accounts of statistics are rare, which is a shame because statistical inference lies at the intersection of philosophy, science, math, and computation, making it a rich and fascinating topic. In this captivating book, Bradley Efron gives a highly readable and entertaining tour of this subject. Without delving into messy technical details, he gives a deep sense of the subtleties and delights that arise when extracting information from data. This is a must-read for anyone who is curious about the way that statistics helps science progress.”—Larry Wasserman, Carnegie Mellon University
“A wonderful opportunity to learn modern statistics from one of its creators, an insightful and patient teacher with a gentle sense of humor.”—Stephen Stigler, University of Chicago
This publication has been produced to meet accepted Accessibility standards and contains various accessibility features including concise image descriptions, a table of contents, a page list to navigate to pages corresponding to the print source version, and elements such as headings for structured navigation. Appearance of the text and page layout can be modified according to the capabilities of the reading system.
Accessibility Features
-
WCAG v2.2
-
WCAG level AA
-
Table of contents navigation
-
Single logical reading order
-
Short alternative textual descriptions
-
Print-equivalent page numbering
-
Landmark navigation
-
Index navigation
-
Epub Accessibility Specification 1.1
-
ARIA roles provided
-
All non-decorative content supports reading without sight
-
No known hazards or warnings
