Theory of Inference Derivation for Qualitative Data: Development and Test withApplication to Criminal and Terrorist Detection

摘要

For this research, the authors developed a process for modeling group decisionmaking. The models are based on analysis of the options selected by members of the group working alone or with others in committees. The selection process is represented as exclusive, exhaustive category schemes (facets) to describe the options available and then the choice of alternatives using these facets. The analysis of these generation and selection processes is modeled using nonmetric multidimensional scaling procedures. A series of studies shows that facet analyses model effectively many aspects of group decision making when the choice criteria involved are essentially qualitative. The application of the facet framework therefore provides a theory of how groups make inference by combining the option selection of their individual members. Aspects such as dominance, alliance, opposition, and involvement, issue salience and relevance and can all be made operational within the facet models of a given group. Studies indicate that by feeding back to a group the results of this type of modeling committee effectiveness may be enhanced. The theory provided a basis for creating an online, interactive group decision support system. Applications of the modeling process to actual decision problems in relation to serial sexual assaults, hijacking, and research grant allocation, as well as student housing selection serve to illustrate the great range of problem solving groups that the theory and a system based on it can assist.