Statistical activity can be divided for descriptive and analytical purposes into (1) discovery/imposition of structure, (2) assessment of variation conditional on structure, and (3) execution of techniques. Each of these three areas of activity has an associated type of uncertainty — structural uncertainty, risk, and technical uncertainty, respectively. In any statistical analysis, an analyst has limited supplies of time, money, know-how and computational power and must use these resources to diminish and to characterize better the three main types of uncertainty and the many subtypes that they include. No existing school of statistical thinking provides a comprehensive framework for considering the various types of uncertainty and the tradeoffs among them that analysts must make. One result of this situation is the absence of a system that properly accounts for all of the types of uncertainty. This Note, reprinted from Statistical Science, v. 2, Aug. 1987, describes the types of uncertainty, catalogues and evaluates current methods for characterizing and diminishing them, considers the types of tradeoffs involved in applying statistical methods, and examines the bias introduced into deliberations because there is no proper system of accounting for uncertainty. The Note is an attempt to begin the construction of such a proper system and thus to reduce or eliminate that bias.
Hodges, James S., Uncertainty, Policy Analysis and Statistics. Santa Monica, CA: RAND Corporation, 1988. https://www.rand.org/pubs/notes/N2793.html. Also available in print form.
Hodges, James S., Uncertainty, Policy Analysis and Statistics, Santa Monica, Calif.: RAND Corporation, N-2793-RC, 1988. As of July 28, 2021: https://www.rand.org/pubs/notes/N2793.html