The need for quantitative measurement represents a unifying bond that links all the physical, biological, and social sciences. Measurements of such disparate phenomena as subatomic masses, uncertainty, information, and human values share common features whose explication is central to the achievement of foundational work in any particular mathematical science as well as for the development of a coherent philosophy of science. This book presents a theory of measurement, one that is "abstract" in that it is concerned with highly general axiomatizations (...) of empirical and qualitative settings and how these can be represented quantitatively. It was inspired by, and represents a generalization and extension of, the last major research work in this field, Foundations of Measurement Vol. I, by Krantz, Luce, Suppes, and Tversky published in 1971. (shrink)

The paper discusses the concept of measurement. Measurement, in the wide sense, is defined as a process of empirical, objective assignment of symbols to attributes of objects and events of the real world, in such a way as to describe them. Strongly defined measurement is measurement that conforms to the paradigm of the physical sciences. Weakly defined measurement is measurement in the wide sense, but which is not strongly defined. Strongly and weakly defined measurements are analysed and compared. Other forms (...) of symbolic representation are distinguished from measurement. (shrink)

The paper introduces what is deemed as the general epistemological problem of measurement: what characterizes measurement with respect to generic evaluation? It also analyzes the fundamental positions that have been maintained about this issue, thus presenting some sketches for a conceptual history of measurement. This characterization, in which three distinct standpoints are recognized, corresponding to a metaphysical, an anti-metaphysical, and relativistic period, allows us to introduce and briefly discuss some general issues on the current epistemological status of measurement science.

Unlike the standard representational theory of measurement, which takes the real numbers as a pregiven numerical domain, the approach presented in this paper is based on an abstract concept of a procedure of measurement, and ‘values of measurement’ are understood in terms of such procedures. The resulting ‘type approach’ makes use of elementary model-theoretic notions and emphasizes the constructibility of scales. It provides a natural starting point for a systematic discussion of issues that tend to be neglected in the standard (...) framework (such as the relation between measurement and computation). At the same time it is perfectly compatible with the modern representational theory of measurement and helps elucidate a number of issues central to that theory (e.g. the role of Archimedean axioms). (shrink)

Ordinary measurement using a standard scale, such as a ruler or a standard set of weights, has two fundamental properties. First, the results are approximate, for example, within 0.1 g. Second, the resulting indistinguishability is transitive, rather than nontransitive, as in the standard psychological comparative judgments without a scale. Qualitative axioms are given for structures having the two properties mentioned. A representation theorem is then proved in terms of upper and lower measures.