(2) how each impact should be measured (e.g., Are distributional effects considered, or just the mean?); and (3) how different outcomes should be weighted.41 For example, risk analysis of nuclear power plants might raise questions about their ability to deliver energy at the promised price, as well as about their potential threats to society. Those risks and benefits may involve human health, the environment, and the economy, as well as the distribution of these risks and benefits, all of which are central societal and ethical concerns. If the ethics of such matters are not explicitly considered, risk analysts are likely to resolve ethical issues by deferring to professional conventions (which are usually based on some ethical framework agreed on in advance) or by imposing their own ethical values and standards.42

Risk analysis seeks to provide a disciplined, transparent way to integrate the knowledge of diverse experts in predicting the performance of a technology in advance of its deployment. It can focus the design process by comparing competing designs and identifying vulnerabilities requiring additional research (e.g., poorly understood properties of materials or social controls on potential uses).43 It can show when the design team lacks critical expertise. It can help decision makers decide whether the benefits of a new technology outweigh its risks, as well as provide the evidence that they need to explain their choices to others.

Risk analyses are soundest when they accommodate a broad range of relevant evidence (e.g., not just readily quantified factors); when they retain awareness of factors that have not been analyzed (e.g., potential design flaws); when they elicit expert judgment with proven methods that are structured to obtain the maximum amount of information from experts; when they do not seek to defend a particular outcome or design or approach; and when they account for uncertainty in the available evidence (e.g., with sensitivity analyses).44 Decision makers need candid


41 Canadian Standards Association, Risk Management Guidelines for Decision Makers, CAN/CSA-850, Ottawa, Ontario, 1997 (reaffirmed 2002); HM Treasury, Managing Risks to the Public: Appraisal Guidance, Her Majesty’s Stationary Office, London, 2005; Sheldon Krimsky and Dominic Golding, Social Theories of Risk, Praeger, New York, 1992.

42 National Research Council, Scientific Review of the Proposed Risk Assessment Bulletin from the Office of Management and Budget, The National Academies Press, Washington, D.C., 2006; Presidential/Congressional Commission on Risk Assessment and Risk Management, Risk Assessment and Risk Management in Regulatory Decision-Making, riskworld.com, 1997, available at http://www.riskworld.com/Nreports/1996/risk_rpt/html/nr6aa001.htm.

43 Baruch Fischhoff, Risk Analysis and Human Behavior, Routledge/Earthscan, Oxford, 2011; Michael S. Wogalter, The Handbook of Warnings, Lawrence Erlbaum Associates, Hillsdale, N.J., 2006.

44 Anthony O’ Hagan, Caitlin E. Buck, Alireza Daneshkhah, et al., Uncertain Judgements: Eliciting Expert Probabilities, John Wiley & Sons, Ltd., Chichester, West Sussex, 2006; E.C. Poulton, Bias in Quantifying Judgment, Lawrence Erlbaum, Hillsdale, N.J., 1989.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement