Statistical measurement originated in repeated observations of one physical quantity with random error.

Evidence

The origins of statistics and measurement lie in the situation where one had multiple measurements of the same physical phenomenon, e.g., the position of a planet. The true position is unobserved, but we do have multiple measurements. The natural thing to do, following the central limit theorem of statistics (and assuming random errors), is to take the mean as the estimate of the position of the planet. — Gary Goertz, Social Science Concepts and Measurement, 351

Categorization

• XREF: Chimes with histories of error theory (Stigler; the ‘average man’ of Quetelet) and connects to the social-science borrowing of physical-science methods.
• DEEPTHINKING: The deep point: latent-variable logic is a transplanted physics procedure whose assumptions of one-thing indicators may not travel to fuzzy social concepts.