FEATURE ARTICLE
The Most Dangerous Equation
Ignorance of how sample size affects statistical variation has created havoc for nearly a millennium
Howard Wainer
For almost a thousand years, there have been documented cases of people drawing erroneous conclusions because of their ignorance of how sample size affects variation of the mean. This was perhaps understandable prior to Abraham de Moivre's derivation of the exact mathematical relation in 1730. Yet the pace of such errors seems only to have increased since then. From the 1160 A.D. trial of the pyx to commentary on Lawrence Summers's recent remarks on sex differences in scientific ability, statistical misunderstandings continue to prevail.
Go to Article