Prescriptive Models for Improving Subjective Probability Judgments

在公共政策和商业环境中，由于专家们对所讨论问题的估计、预测甚至性质的分歧，往往使决策复杂化。此外，面对这样的分歧，做出决定往往是必要的；在一段合理的时间内收集所需的“确凿证据”可能根本不可行。因此，找到利用专家所拥有的宝贵知识的方法是很重要的。专家知识通常被最好地描述为“意见”，可以用概率判断来表达，即专家对某事为真可能性的评估。使用专家概率判断的典型公共政策设置包括能源（特别是核电和工程）、军事、情报和安全、环境和卫生政策以及经济预测。在商业环境中，专家判断在总体战略决策和石油、天然气勘探和制药研发等情况下发挥着重要作用。对于提高专家的概率判断，没有简单的答案。心理学研究已经在很多方面证明了这种判断是有偏见的。因为不同的心理过程可能在不同的情况下起作用，所以抵消相应偏见的方法必须适应特定的情况。提出了提高专家概率的三种实用方法。这三种方法都以心理学理论为基础，设计用于各种概率评估情况。每种方法都将经过严格的评估以证明其有效性。这三种方法进一步为科学家如何利用心理学结果和理论来开发额外的调整方法以提高专家概率提供了模型。

Decision making in public policy as well as commercial settings is oftencomplicated by disagreement among experts regarding estimates,forecasts, and even the nature of the problem under discussion.Moreover, it is often necessary to make a decision in the face of suchdisagreements; collecting the required "hard evidence" may simply beinfeasible within a reasonable period of time. Thus, it is important tofind ways to use the valuable knowledge that experts possess. Expertknowledge often is best characterized as "opinions" that can beexpressed as probability judgments, the expert's assessment of thechance that something is true. Typical public-policy settings whereexpert probability judgments are used include energy (especially nuclearpower and engineering), military, intelligence and security, environmentand health policy, and economic forecasting. In commercial settings,expert judgments play large roles in overall strategic decision makingand situations such as oil and gas exploration and pharmaceuticalresearch and development.No simple answer exists for improving experts' probability judgments.Psychological research has documented many ways in which such judgmentsare biased. Because different psychological processes may be at work indifferent situations, methods to counteract the corresponding biasesmust be adapted to the specific situation. The proposed research offersthree practical methods for improving experts' probabilities. All threemethods are grounded in psychological theory and are designed to beapplicable to a variety of probability-assessment situations. Eachmethod will be subjected to rigorous evaluation to demonstrate itseffectiveness. The three methods further provide models for how ascientist can use psychological results and theory for developingadditional adjustment methods for improving expert probabilities.