247
measurement is an empirical and informational process, designed on purpose, whose input
is an empirical property of an object and that produces information in the form of values of
that property
Chapter 3 added three key specifications to this standpoint. First, though sometimes
neglected in nonscientific situations, measurement results should include information about the quality of the reported values, which in the past was described in
reference to measurement errors but is today more usually modeled in terms of
uncertainty and validity, in physical and psychosocial measurement, respectively.
Second, measured values report a relational form of information—the ratio of
the measured property to the chosen unit, in quantitative cases—as inherited from
the Euclidean tradition: this requires the social availability of a metrological system
aimed at disseminating the reference properties by means of measurement standards
mutually connected in traceability chains. Hence measurement requires calibration,
and measurement standards make the calibration of measuring instruments possible.
Third, despite its historical importance, the actual relevance of the Euclidean
tradition to measurement science has been overemphasized: indeed, it refers to the
mathematical concept, i.e., a number as a ratio of entities, which is only
loosely related to the abovementioned empirical and informational process of measurement. The point is then that the condition that measurement applies only to
quantitative properties cannot be justified by reference to the Euclidean tradition.
The rest of the book can be interpreted as a report of our explorations around one
question: Given these necessary conditions, what complementary conditions are
sufficient to characterize measurement?
With that in mind, we discussed, in Chap. 4, the epistemic status of measurement
and the conditions of its proper use, as understood in the context of the three broad
perspectives of realism, operationalism, and representationalism. The main findings
were presented in a simple two-by-two matrix whose dimensions specify whether
measurement has been characterized as being dependent on empirical and/or mathematical constraints, respectively, which led to the conclusion that what characterizes measurement is the empirical structure of the process, not mathematical
constraints on the inputs or the outputs of the process. This is in fact the position that
we have developed, coupled with the acknowledgment of the unavoidability of the
role of models in the process, thus grounded by what could be called a modeldependent realism about measurement.
Not surprisingly, the next stage of the exploration was about the very target of
measurement, i.e., properties, which were analyzed in Chap. 5 from both ontological and epistemological perspectives. Here the core issue is as simple as it is controversial, in that it concerns the actual meaning of the Basic Evaluation Equation,
property of an object value of a property
=
which is the basic structure of any measurement result (and which must also be
complemented with information about uncertainty). Consistent with our modeldependent realist standpoint, we interpreted this relation as the claim of an actual
referential equality: it conveys information on the measurand because the measur8.2 The path we have walked so far
measurement is an empirical and informational process, designed on purpose, whose input
is an empirical property of an object and that produces information in the form of values of
that property
Chapter 3 added three key specifications to this standpoint. First, though sometimes
neglected in nonscientific situations, measurement results should include information about the quality of the reported values, which in the past was described in
reference to measurement errors but is today more usually modeled in terms of
uncertainty and validity, in physical and psychosocial measurement, respectively.
Second, measured values report a relational form of information—the ratio of
the measured property to the chosen unit, in quantitative cases—as inherited from
the Euclidean tradition: this requires the social availability of a metrological system
aimed at disseminating the reference properties by means of measurement standards
mutually connected in traceability chains. Hence measurement requires calibration,
and measurement standards make the calibration of measuring instruments possible.
Third, despite its historical importance, the actual relevance of the Euclidean
tradition to measurement science has been overemphasized: indeed, it refers to the
mathematical concept
loosely related to the abovementioned empirical and informational process of measurement. The point is then that the condition that measurement applies only to
quantitative properties cannot be justified by reference to the Euclidean tradition.
The rest of the book can be interpreted as a report of our explorations around one
question: Given these necessary conditions, what complementary conditions are
sufficient to characterize measurement?
With that in mind, we discussed, in Chap. 4, the epistemic status of measurement
and the conditions of its proper use, as understood in the context of the three broad
perspectives of realism, operationalism, and representationalism. The main findings
were presented in a simple two-by-two matrix whose dimensions specify whether
measurement has been characterized as being dependent on empirical and/or mathematical constraints, respectively, which led to the conclusion that what characterizes measurement is the empirical structure of the process, not mathematical
constraints on the inputs or the outputs of the process. This is in fact the position that
we have developed, coupled with the acknowledgment of the unavoidability of the
role of models in the process, thus grounded by what could be called a modeldependent realism about measurement.
Not surprisingly, the next stage of the exploration was about the very target of
measurement, i.e., properties, which were analyzed in Chap. 5 from both ontological and epistemological perspectives. Here the core issue is as simple as it is controversial, in that it concerns the actual meaning of the Basic Evaluation Equation,
property of an object value of a property
=
which is the basic structure of any measurement result (and which must also be
complemented with information about uncertainty). Consistent with our modeldependent realist standpoint, we interpreted this relation as the claim of an actual
referential equality: it conveys information on the measurand because the measur8.2 The path we have walked so far