26
Of course, this condition is not sufficient to characterize measurement: there are
plenty of input-output empirical processes that are not measurements. An example
is combustion, in which oxygen is chemically combined with other substances
(inputs) and heat and light are produced (outputs). Clearly combustion is not
measurement. However, we propose that all examples of measurements can be
uncontroversially characterized as input-output empirical processes.
In the simple case of the measurement of human body temperature by means of
a mercury thermometer, the basic input is the temperature to be measured and the
basic output is the information obtained on this temperature, in the form of a value
such as 36.8 °C. More generally, other inputs could contribute to the output, including the temperature and the atmospheric pressure of the environment; an intermediate output is the final position of the mercury in the glass tube of the thermometer
as the result of the interaction between the individual and the thermometer itself,
and a final output of the measurement would be the temperature in degrees Celsius,
whereas the final outcome of the whole process could be the assessment whether the
considered human being is sick or in good health.
11
An example in the human science domain is the measurement of a student’s
reading comprehension ability, which illustrates some of the complexities that can
underlie the concept of empirical input-output process. A typical scenario for the
measurement of reading comprehension ability was given in Sect. 1.2.2, and might
result in an estimate of the student’s reading comprehension ability in the form of a
value on a reading comprehension scale, and a teacher might use interpretational
materials to assign the student to a category that involves some specific reading
instruction activities. The basic input here is the student’s comprehension of the raw
text, and the basic output is the estimated value of the student’s reading comprehension ability. As in the case of temperature, other inputs are usually present that could
contribute to the output, such as distracting noises and mood of the student; an
intermediate output is the set of student responses to the comprehension questions,
and a final outcome could be seen as the category of instructional activities assigned
by the teacher.
2.2.2 Measurement as a designed process
The second black box condition of measurement is that it is a process designed on
purpose, rather than a transformation that spontaneously happens. This has the consequence that measurement is performed according to specifications, called a measurement procedure, as depicted in Fig. 2.3.
11 If the final output of the measurement process were the decision about the person’s health state
rather than the temperature value, then this would be an example of a nonquantitative evaluation,
whose values might be, e.g., healthy, rather sick, and seriously sick. We return to a discussion of
the conditions under which such an evaluation might be considered a measurement in Sect. 6.5.
2 Fundamental concepts in measurement
Of course, this condition is not sufficient to characterize measurement: there are
plenty of input-output empirical processes that are not measurements. An example
is combustion, in which oxygen is chemically combined with other substances
(inputs) and heat and light are produced (outputs). Clearly combustion is not
measurement. However, we propose that all examples of measurements can be
uncontroversially characterized as input-output empirical processes.
In the simple case of the measurement of human body temperature by means of
a mercury thermometer, the basic input is the temperature to be measured and the
basic output is the information obtained on this temperature, in the form of a value
such as 36.8 °C. More generally, other inputs could contribute to the output, including the temperature and the atmospheric pressure of the environment; an intermediate output is the final position of the mercury in the glass tube of the thermometer
as the result of the interaction between the individual and the thermometer itself,
and a final output of the measurement would be the temperature in degrees Celsius,
whereas the final outcome of the whole process could be the assessment whether the
considered human being is sick or in good health.
11
An example in the human science domain is the measurement of a student’s
reading comprehension ability, which illustrates some of the complexities that can
underlie the concept of empirical input-output process. A typical scenario for the
measurement of reading comprehension ability was given in Sect. 1.2.2, and might
result in an estimate of the student’s reading comprehension ability in the form of a
value on a reading comprehension scale, and a teacher might use interpretational
materials to assign the student to a category that involves some specific reading
instruction activities. The basic input here is the student’s comprehension of the raw
text, and the basic output is the estimated value of the student’s reading comprehension ability. As in the case of temperature, other inputs are usually present that could
contribute to the output, such as distracting noises and mood of the student; an
intermediate output is the set of student responses to the comprehension questions,
and a final outcome could be seen as the category of instructional activities assigned
by the teacher.
2.2.2 Measurement as a designed process
The second black box condition of measurement is that it is a process designed on
purpose, rather than a transformation that spontaneously happens. This has the consequence that measurement is performed according to specifications, called a measurement procedure, as depicted in Fig. 2.3.
11 If the final output of the measurement process were the decision about the person’s health state
rather than the temperature value, then this would be an example of a nonquantitative evaluation,
whose values might be, e.g., healthy, rather sick, and seriously sick. We return to a discussion of
the conditions under which such an evaluation might be considered a measurement in Sect. 6.5.
2 Fundamental concepts in measurement