73
of measurement, or scientific inquiry more generally, which is in general expected
to produce objective information, i.e., to be epistemically objective.
The tension between objectivity and subjectivity in measurement can even be
regarded as the basic philosophical problem of measurement (Mari, 2003): What
distinguishes measurement from generic evaluation?
1
In the course of history it has been suggested that measurement can be characterized in reference to:
2
• Ontic reasons: e.g., measurement is a process designed to discover the values
that properties of objects inherently have.
• Epistemic reasons: e.g., measurement is a process designed to produce true, or at
least credible, information on the properties intended to be measured.
• Pragmatic reasons: e.g., measurement is a process designed to be adequate for its
goal of obtaining information on the properties intended to be measured.
• Formal reasons: e.g., measurement is a process designed to evaluate properties in
a consistent way by means of symbols.
These positions are not mutually exclusive. It seems fair to say that any wellconceived standpoint on measurement would acknowledge the validity of more than
one (and possibly all) of these positions, though there may be room for differences
in terms of the balance of their relative importance. On this basis, the next section is
devoted to exploring some of the most influential perspectives on these issues, and
in particular to how each of them leads to different characterizations of the necessary and sufficient conditions for measurement.
4.2 Characterizing measurement
A wide range of characterizations of measurement are found throughout the scientific and technical literature, witnessing both the wide interest in the subject and its
complexity. This multiplicity can be interpreted according to some complementary
criteria (Mari, 2013), as in Table 4.1.
For each criterion, the incompatibility of the two positions is manifest. And
while these incompatibilities have not prevented important advancements in measurement science, such a confused and confusing situation is clearly not optimal.
We propose that one can understand this multiplicity as being the result of some
stereotypes that affect measurement (Mari, Maul, Torres Irribarra, & Wilson, 2017):
each of these positions is grounded on some reasonable basis and many are attractive for their simplicity, but each of the interpretations they provide misses some key
aspects of the complex concept of measurement. In this section we analyze those
1 As noted in Footnote 19 of Chap. 2, we use the term “evaluation” to refer to an attribution of a
value to the property of an object.
2 For an introduction to the philosophical understanding of measurement see also Tal (2020).
4.2 Characterizing measurement
Précédent

- 107/319

Suivant