Originally published in 1989, this volume carefully catalogues and classifies the effect of bias on quantitative judgments. Each bias is described by model and examples of this bias are given. The examples include both theoretical investigations carried out in the laboratory to study bias and practical illustrations. They are accompanied by a discussion of methods of avoiding bias, where this is possible. An early look into this now widely recognised subject, this book was an invaluable resource for investigators and practitioners in applied science at the time.

This book discusses the well-known fallacies of behavioral decision theory. It shows that while an investigator is studying a fallacy, he or she may introduce without realizing it, one of the simple biases that are found in quantifying judgments. The work covers such fallacies as the apparent overconfidence that people show when they judge the probability of correctness of their answers to two-choice general knowledge questions using a one-sided rating scale; the apparent overconfidence in setting uncertainty bounds on unknown quantities when using the fractile method; the interactions between hindsight and memory; the belief that small samples are as reliable and as representative as are large samples; the conjunction fallacy for Linda and Bill; the causal conjunction fallacy; the regression fallacy in prediction; the neglect of the base rate in the Cab problem, in predicting professions, and in the Medical Diagnosis problem; the availability and simulation fallacies; the anchoring and adjustment biases; Prospect theory; and bias by frames. The aim of this book is to help readers to learn about the fallacies and to avoid them.