Abstract
In the last few decades foundations of measurement have developed so as to account for both the role of modeling in measurement, in particular relating to the presence and the effects of measurement uncertainty, and the fact that any measurement is performed by using instruments that work on the basis of transduction effects and provide justified results only in so far as they are properly calibrated. This has triggered a new interest about the role of instruments in the models of measurement. The structure of the process has been variously studied in reference to the connection between measured properties and indications provided by instruments, and to the way in which intersubjective information on the measurand is acquired through instrument calibration. From such a background this paper proposes a comprehensive structural model of direct measurement, whose functional elements, empirical and informational, are presented with a bottom-up strategy as a set of interrelated modules. The result, shown to be a generalization of some of the models currently available in the literature of measurement science, highlights the key role of scales for measurement, clarifies the conceptual and operative relations between direct measurement and calibration, and identifies the principal sources of measurement uncertainty in the structural context of the process. This model is intended to interpret both physical and non-physical measurements, and as such it is a component of a "measurement across the sciences" research programme.