Influence of Attribute Formulation on the Evaluation of Apartments by Multi-Attribute Utility Procedures

Abstract

As we have already learned from the contributions of several participants of this conference (e. g., Bauer & Wegener’s comments on Humphreys’ paper), one of the main problems in the application of multi-attribute utility (MAU) theory to real world decision problems seems to be the attributes per se. Actually, this is not a problem inherent in the theory per se, since MAU theory, or at least the model testing part of it, might be applied to any attributes. The usefulness of the results of MAU analysis will, however, depend to a great extent on the attributes used and their relation to one’s objectives. Now usually these attributes have to be searched for, identified formulated, and operationalized before we can start to evaluate anything. This importance of this structuring phase or goal analysis is stressed, for example, by the number of pages devoted to this problem in such basic texts as Raiffa (1969) or Miller III (1970). These attribute formulation procedures have many of the influences of a creative process and as such are subject to some random influences. The results, in our case the attributes, are not only determined by the goals of the decision maker, but are also influenced by many situational and personal factors more or less outside the control of the decision analyst.