Theory and measurement: Why representation is not enough

Measurement in psychology generally aims to assign numbers to objects, such that the numbers represent one or more psychological attributes. Representational measurement theory provides a framework to test whether such a representation (e.g., the mapping of subjective magnitudes, such as perceived brightness, to the real numbers) is possible in the first place. However, representational measurement does not acknowledge the pivotal role of substantive formal theory for model-building and measurement. This chapter focuses on the role of scientific theory construction for the practice of measurement in psychology and other sciences. I argue that substantive formal theory is necessary to bridge the gap between representations of supposed psychological attributes and their theoretical meaning. Lacking a sound foundation in substantive formal theory, current psychological testing procedures may serve a social purpose (e.g., in personnel selection or educational benchmarking), but do not advance the understanding of human behavior on a scientific level.