cult of statistical significance

‘Cult of Statistical Significance’ quoted on Slate; elsewhere, the authors continue the discussion

by Shaun Manning October 5, 2012

In an article entitled “The Internet Blowhard’s Favorite Phrase” (also published as “Stop Saying That Correlation Does Not Imply Causation”), Slate writer Daniel Engber ponders how “a stats-class admonition become so misused and so widespread” as “the statistical cliché that closes threads and ends debates.” “No, correlation does not imply causation, but it sure as hell provides a hint,” Engber said. Citing a recent study linking internet usage to depression–which, predictably, elicited the correlation/causation response online–Engber explained that that the lack of a definitive cause-effect relationship does not make the observation meaningless. “Does email make a man depressed? Does sadness make […]

Read more

Freakonomics and WaPo blogs pick up Ziliak’s Guinness study

by Shaun Manning February 13, 2012

Statistics and beer continue to be a winning combination, as the popular Freakonomics blog and Ezra Klein’s Wonkblog at the Washington Post picked up the story of Stephen T. Ziliak’s recent paper on the study of Guinness in the Journal of Wine Economics. Ziliak, co-author with Deirdre N. McCloskey of The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives, has been an outspoken critic of research methods that elevate the importance of statistical significance over more subjective–but often more relevent–factors such as the magnitude of an effect or the quality produced. “Gosset (1876–1937) aka […]

Read more

Beer and Stats: Chicago Magazine covers Ziliak’s Guinness Experiment

by Shaun Manning February 10, 2012

Stephen T. Ziliak, co-author with Deirdre n. McCloskey of The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives, is the subject of a Chicago Magazine blog post on “Guinness beer and Guinnessometrics.” The Chicago piece summarizes Ziliak’s paper in the Journal of Wine Economics, which focuses on the work on an early 20th-century chemist-turned-brewer at Guinness. After a discussion of the experiments undertaken by William Sealy Gosset–aka “Student”–testing the three main ingredients for stout, Ziliak comes around to a conclusion that is very much in keeping with the argument of U-M Press book, that […]

Read more

Citing Recent Court Cases, Professor Praises U-M Press Title

by Shaun Manning December 6, 2011

In a recent blog post considering high court decisions in the United States and Britain, Michael Smithson, Professor of Psychology and Decision Sciences at Australia National University, called Stephen T. Ziliak and Deidre N. McCloskey’s Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives “a swinging demolition of the unquestioned application of statistical significance in a variety of domains.” Like Ziliak and McCloskey, Dr. Smithson notes that there are meaningful applications of statistical significance, but context is important–what is being measured, what is the sample size, and what a non-significance finding might actually mean are […]

Read more