Measurement is a ubiquitous feature of science, engineering, commerce and everyday life. It is often considered a hallmark of the scientific enterprise and a privileged source of knowledge relative to qualitative modes of inquiry. Despite its widespread use, there is little consensus among philosophers on what precisely constitutes measurement or what sorts of things are measurable. This article outlines several philosophical approaches to the subject and examines the problems that arise in each.
Measurement theory what is measurement is a highly heterogeneous body of work that spans from the nineteenth century to the present and endorses a wide range of views about the metaphysics, epistemology and semantics of the concept. Among the most important issues are the nature of the relata (objects) whose relations numbers mirror, and how those objects relate to one another.
Most theories of measurement focus on the process by which a physical quantity is translated into a number or other mathematical representation. Such translations are usually characterized by a set of rules and procedures, which are known as a measurement system.
A key part of a measurement system is the choice of units, which are defined on a scientific basis and overseen by governmental or independent agencies such as the General Conference on Weights and Measures. Historically, various systems of units have been used for different purposes, but developments in the nineteenth century led to the adoption of a unified standard for metric measurement (the International System of Units) that reduces all physical measurements to a mathematical combination of seven base units.
There are many different kinds of measurement, ranging from the purely descriptive to the objective and abstract. The latter include time, frequency and distance, while the former encompasses the properties of space and motion. Regardless of the type of measurement, all require some sort of signaling device, which identifies an object or phenomenon to be measured, and transmits a physical signal to it. The signaling device may be powered by the object itself, or it may be triggered by interaction with it by means of a physical probe.
The resulting information is compared to a reference signal of known quantity, and the difference between the two forms of signaling is calculated. The error incurred during the measurement process is an essential part of the problem of measurement and must be taken into account by any theoretical approach.
Some of the major theories of measurement are based on different metaphysical and epistemological positions, such as realism, conventionalism, operationalism, model-based accounts and information-theoretic approaches. The philosophies of measurement also differ in their perspectives on the structure of quantity terms and the metaphysical status of measurable quantities. Nonetheless, all of these theories have at least some overlap with each other and share some features of the general problem of measurement. The more narrow characterization of measurement as the correlation of numbers with entities that are not numbers, which is implicit in the work of Campbell, may be too broad to qualify as a useful definition for some scientific experiments.