Search

This study investigated the goodness of one behavioral and five mathematical aggregation rules for combining likelihood ratio estimates given by a group of experts. Behavioral aggregation required only that all four members of the group agree to use a single estimate. All mathematical aggregation rules used the geometric mean of the estimates. However, five methods were used to weight the estimates of each "expert." The methods used were equal weights, peer-assigned weights, self-assigned weights, group (combined peer and self) assigned weights and an average between self- and group-assigned weights. Statistical differences were found in favor of mathematical consensus for likelihood ratios from size 2 to 15. No differences were found for smaller or larger likelihood ratios, or between the various weighting methods. Equal weighting of experts was the simplest method to use. Estimates were obtained at a rate of 3.25 mm per group per estimate. [ABSTRACT FROM AUTHOR]

This research evaluated four methods of eliciting subjective likelihood ratio estimates. The methods differed in terms of amount and structure of interaction permitted between estimators. These processes were individual estimates, and three group processes: a Talk-Estimate process approximating an interacting group, an Estimate-Feedback-Estimate process as an approximation of a Delphi group, an Estimate-Talk-Estimate process as combination of nominal and interacting groups. In this study the Estimate-Talk-Estimate group process was superior in approaching correct estimates in this judgmental task. This is consistent with the long research tradition which favors group as opposed to individual problem-solving in judgmental situations. The individual Estimate process and the Estimate-Feedback-Estimate technique performed about equally well with respect to both error and variability. If anything, written feedback appeared to lead to a reduction in the quality of estimates. Finally, the relatively poor results from the Talk-Estimate process are consistent with other studies which have pointed out dysfunctions of inter- acting group processes for judgmental tasks. [ABSTRACT FROM AUTHOR]

A Bayesian probabilistic information processing system (PIP) was compared, in field and laboratory tests, with four other forecasting methodologies. The influence of amount of information and conditional dependence of data on forecaster performance was investigated. PIP performed at least as well as any other technique in terms of magnitude of error and was superior to all techniques in terms of frequency of error. However PIP also required the most physician time per estimate. [ABSTRACT FROM AUTHOR]