Multi-attribute evaluation processes: Methodological and conceptual issues

Multi-attribute evaluation situations are characterised by a judge, a set of alternatives and a set of attributes. Each alternative has a value on each attribute, and the judge's cognitive process leading to an evaluation of the alternatives on the basis of their attribute values is a major topic in psychological judgement and decision-making research. In this article we focus on methodological and conceptual issues related to this research. First we elaborate on two types of multi-attribute evaluation, judgement and choice. Next, we discuss the development and characteristics of the research paradigms used for studying multi-attribute judgement and choice: structural modelling and process tracing respectively. A central issue in both paradigms is the question whether the multi-attribute evaluation process follows compensatory or noncompensatory principles. We relate this issue to various aspects of the research paradigms (linearity, additivity and configurality of the models used in the structural modelling approach, and information load and response mode in the process-tracing approach), and conclude that, in general, judgement follows compensatory and choice noncompensatory principles. Finally, we discuss conceptual and methodological problems associated with the two research paradigms, and mention new directions for future research.