Thomas Stewart - Academia.edu (original) (raw)

Papers by Thomas Stewart

Journal of Forecasting, Dec 1, 1994

S cientific research on atmospheric processes and weather forecasting frequently leads to improve... more S cientific research on atmospheric processes and weather forecasting frequently leads to improvements in weather forecasts [such as quantitative precipitation forecasts (QPF)], as measured by scientific criteria. Furthermore, effective use of such improvements has the potential for avoiding injury and death, averting property and environmental damage, and other societal benefits. Although the potential societal value of improved weather forecasts is substantial , the realization of that potential is not automatic (Pielke and Carbone 2002). The scientific community has a responsibility to work toward that realization. The success of that work depends on a number of factors. Obtaining the actual (not just the potential or theoretical) value of improved forecast technology requires (a) a forecasting process that translates improved science and technology into improved forecast products that are targeted to user needs, (b) effective communication of forecast information to users in a timely fashion and in a form useful for making weather-information-sensitive decisions, and (c) users who incorporate the forecast product into their decisions in order to make better choices (on the processes of forecasts, communication, and use, see Glantz and Tarleton 1991; Piekle and Carbone 2002; Sarewitz et al. 2002). Meeting these requirements begins with the detailed understanding of user needs and decision processes that is the outcome of systematic study. Such study also provides a foundation for studies of the value of current forecasts and forecast improvement. There is a large literature on forecast value (see, e.g., UNDERSTANDING USER DECISION MAKING AND THE VALUE OF IMPROVED PRECIPITATION FORECASTS

Weather and Forecasting, Mar 1, 1989

This study compared meteorologists, an expert system, and simple weighted-sum models in a limited... more This study compared meteorologists, an expert system, and simple weighted-sum models in a limited-information hail forecasting experiment. It was found that forecasts made by meteorologists were closely approximated by an additive model, and that the model captured most of their forecasting skill. Furthermore, the additive model approximated the meteorologists' forecasts better than the expert system did. Results of this study are consistent with the results of extensive psychological research on judgment and decision making processes. Potential implications are discussed.

constructivist coding hypothesis to describe how people code the outcomes of their decisions when... more constructivist coding hypothesis to describe how people code the outcomes of their decisions when availability of feedback is conditional on the decision. They provided empirical evidence only for the.5 base rate condition. This commentary argues that the constructivist coding hypothesis imposes an ever-declining selection rate and overestimates base rate bias for high base rate conditions. We provide support based on a simulation model of learning under selective feedback with different base rates. Then we discuss possible extensions to constructivist coding that can help overcome the problem.

Organizational Behavior and Human Performance, 1975

When subjective descriptions of judgmental policies have been compared to statistical description... more When subjective descriptions of judgmental policies have been compared to statistical descriptions of policies, investigators have repeatedly found an imperfect correspondence between subjective and statistical measures of importance. There is reason to believe, however, that these findings may be due to the use of inadequate procedures for obtaining subjective weights. The present study, therefore, compared seven methods for obtaining subjective descriptions of cue importance. Subjective policy descriptions were found to correspond reasonably well to statistical policy descriptions for both a three-cue and a seven-cue task, but the seven methods did not dlffer. All methods yielded more accurate predictions of actual judgments than an arbitrary policy of equal weights on all cues.

Journal of the American Institute of Planners, 1976

Adaptive Perspectives on Human–Technology InteractionMethods and Models for Cognitive Engineering and Human-Computer Interaction, 2009

William J. Horrey, Christopher D. Wickens, Richard Strauss, Alex Kirlik, and Thomas R. Stewart 5 ... more William J. Horrey, Christopher D. Wickens, Richard Strauss, Alex Kirlik, and Thomas R. Stewart 5 Supporting Situation Assessment through Attention Guidance and Diagnostic Aiding: The Benefits and Costs of Display Enhancement on Judgment Skill Introduction Many operational ...

As the U.S. National Weather Service modernizes its forecast offices in the 1990s. new technologi... more As the U.S. National Weather Service modernizes its forecast offices in the 1990s. new technologies will vastly increase the volume of data available to the forecaster. We anticipate that knowledge-based computer systems will aid forecasters in two important ways: by guiding forecasters to data pertinent to the forecast problem of the day. and by providing infomation to forecasters faster and more efficiently than before. Shootout-91 compared howledge-based computer systems designed to forecast significant and severe convective-season weather in Oklahoma and eastern Colorado. Goals of this experiment were to 1) explore the ability of knowledge-based systems to make forecasts of severe and significant weather, 2) study human-computer interaction. and 3) compare different forecast techniques. In this paper we repon on the analysis of the forecasts made by the systems and the forecasters. discuss forecasters' experiences with the systems, and present our vision of the future of howledgebased systems in the forecast office. Comme le service national mbk5orologique des Etats-Unis modernise ses installations dans cette decade 90, les nouvelles technologies vont augmenter le volume de d 0~6 e S dqmnible pwr le m6t&rologiste. C'est pour cela par anticipation nous disons que la connaissance de base du systeme infomatique aidera le.met&rologiste dans deux voie aeS importantes qui: 1) Guideront le met&rologiste directement vers les dom&s pettinentes du probleme de pMictabilit6 du jow 2) Fourniront au methologiste des informations d'une manibre plus rapide et plus 6fficace qu'auparavant. Le rappon de confrontation 91 compare d'une maniere significative le choix du systeme de base d'infomatique applique dans la prevision, et plusieurs temps saisoniques dB la convection armospherique dans I'Oklahma et I'est du Colorado. Le but de ces experiences elaient de: I) Explorer L'abilite du systeme de connaissance de base pour faire des pdvisions d'une maniCre significative & plusiem temps: 2) Etudier I'iteraction ordinate&em humain. 3) Enfm companr d i f f h l e s techniques de @vision. Dans ce papier nous rapponons I'analyse de pdvisions faite par le syseme et le m6k5orologiste. la discussion des exp6riences du m 6 h l o g i s t e avec le systeme; et nous pdsentons now vision du future de la connrussance de base du systeme informatiquc dans les services & @visions.

Medical Decision Making, 1988

A method for increasing the statistical power of clinical trials to detect clinically important d... more A method for increasing the statistical power of clinical trials to detect clinically important differences is described. Inconsistency in physicians' overall judgments of treatment effec tiveness adds "noise" to a trial that may mask either the superiority or the inferiority of particular treatments. The method described here uses "judgment analysis" to reduce errors in an individual's overall judgments of treatment effectiveness. The method can also be used to reduce error variance due to differences in judgment between physicians and may thus be particularly useful in multicenter trials. The method is illustrated with results from a recent trial. Key words: statistical power; experimental design; multi-investigator trials; judgment analysis; clinical judgment. (Med Decis Making 8:33-38, 1988)

Weather and Forecasting, 1990

Estimates of several components of forecasting skill can be obtained by combining a skill-score d... more Estimates of several components of forecasting skill can be obtained by combining a skill-score decomposition developed by Allan Murphy with techniques for decomposing correlation coefficients that have been employed in research on human judgment. The decomposition of the correlation coefficient requires knowledge of the information or "cues" used by the forecaster. When the cues are known, it is possible to estimate the effects of uncertainty and the forecaster's consistency and use of the cues.

Risk Analysis, 1986

The risk of oil spills is a major environmental issue in the siting of proposed coastal refinerie... more The risk of oil spills is a major environmental issue in the siting of proposed coastal refineries, oil terminals, deepwater ports, and in the leasing of offshore lands for oil exploration and development. As with any kind of risk, oil spill risk assessment is inherently judgmental and no analytic method can eliminate the need for judgment. This paper compares representative examples of oil spill risk assessments with regard to decisions about data, variables, functional relations, and uncertainty. The comparison emphasizcs the judgmental basis of analytic methods.

Medical Decision Making, 2003

Purpose. To understand why many primary care physicians in the United States and France order pro... more Purpose. To understand why many primary care physicians in the United States and France order prostate-specific antigen (PSA) tests routinely for their asymptomatic male patients despite “evidence-based” recommendations. Methods. Thirty-two U.S. general internists and family practitioners and 33 French generalists judged, for 32 hypothetical male patients seen for routine preventive care, the probability that the patients had asymptomatic prostate cancer and the likelihood that they would order PSA tests. They were also asked about beliefs, attitudes, and knowledge related to prostate cancer. Results. The significant predictors of ordering more PSA tests in the scenarios were physicians' higher ratings of regret if untested patients were found to have advanced cancer, their greater discomfort if they suspected that patients had illnesses but could not know for sure, and their perceptions of official recommendations as favoring routine testing. Implications. Nonrational factors c...

International Journal of Forecasting, 1999

Environment and Behavior, 1988

Farmers of the Ogallala Aquifer region of the Great Plains in the western United States were inte... more Farmers of the Ogallala Aquifer region of the Great Plains in the western United States were interviewed concerning their perceptions of drought, weather and climate changes, aquifer conditions, and the adaptive strategies used to cope with these. The results of these surveys parallel and refine the results reported by Saarinen in 1966: Farmers recall classic droughts and the most recent years as drought; intermediate years and droughts are lost from memory. A pattern of perception is suggested by these results wherein experience influences both definition and memory, which in turn affect expectations of the particular environmental event. People's perceptions, expectations, and adaptive responses to recurrent environmental stimulae, such as drought, are related to their direct experiences. When one's experience restricts his or her view of the potential variance in meteorological or other environmental events, however, that restriction may lead to inappropriate or insuffici...

Prediction: Science, Decision Making, and the …, 2000

Behavioral and Brain Sciences, 2014

We elucidate the epistemological futility of using concepts such as unconscious thinking in resea... more We elucidate the epistemological futility of using concepts such as unconscious thinking in research. Focusing on Newell & Shanks' (N&S's) use of the lens model as a framework, we clarify issues with regard to unconscious-thought theory (UTT) and self-insight studies. We examine these key points: Brunswikian psychology is absent in UTT; research on self-insight did not emerge to explore the unconscious; the accuracy of judgments does not necessitate the unconscious; and the prescriptive claim of UTT is unfounded.

The American Journal of Psychology, 1974

The 80 subjects had a task of a two-cue probability learning at a +.8, +.4, .O,-.4, or-.8 level o... more The 80 subjects had a task of a two-cue probability learning at a +.8, +.4, .O,-.4, or-.8 level of stimulus redundancy. An analysis of 50 lcarning trials revealed a significant main cffect of redundancy. Performance was higher whcn redundancy was lower. Thcse findings are interpreted in terms of task characteristics that accompany redundancy, and the implications for the learning of complex inference tasks are discussed.

The Oxford Handbook of Economic Forecasting, 2011

This article, which deals with methods for quantifying the economic value of weather and climate ... more This article, which deals with methods for quantifying the economic value of weather and climate forecasts, is organized as follows. Section 2 provides some background on methods used to produce weather and climate forecasts, including the distinction between “weather” and “climate.” Section 3 introduces the concept of the economic value of imperfect information, based on the framework of decision theory and expected utility maximization. Section 4 reviews specific decision-analytic studies of the economic value of weather and climate forecasts. As a complement to the decision-theoretic approach, nonmarket valuation of weather and climate forecasts based on stated preference methods are described in Section 5. As an example, a recent survey of the public to obtain willingness-to-pay estimates for the economic value of improved hurricane forecasts is treated in detail. Finally, Section 6 consists of a discussion focusing on future research directions that could result in improved ass...