Cognitive Prerequisites for Scaleable Data: A Model of Impression Formation

Abstract

In this chapter, formal requirements are specified which must hold for subject responses in attitudinal assessment or person perception, in order to satisfy certain levels of measurement. Ordinal, interval, and ratio scale levels are considered. Producing individual data on each of these levels is conceptualized as a set of cognitive tasks of increasing complexity. Models of these tasks are formalized as rules. Furthermore, the perceiver (or, the judging person) is assumed to possess a target-independent preference order with respect to different traits or features that might be observed in a person or object. Using pair comparison as the underlying method of data collection, the chapter shows how given subsets of traits will lead to evaluations about some target, and how comparisons between different targets can be made. Depending on the particular rules involved in these processes, the resulting comparison data may, across repeated judgments, satisfy a particular scale level, allowing to draw conclusions about the rules underlying those judgments. The effects of context information on target comparisons are discussed, and examples are introduced for all specified rules and scale levels. The general argument is made that formal measurement theory is a useful tool for impression formation and attitude research.