This is the beginning of a thread in which I, intermittently and when I remember, add quotes and comments from a book I’m reading called “Measurement in Psychology “ by Joel Michell ...

“Unless every attribute really is quantitative, to conclude that, because one can make numerical assignments to things, the attribute involved must be measurable, is to presume upon nature.”

Michell notes that in other (ie physical) sciences, an attribute is quantitative & thus measurable if it is structured such that diff magnitudes of it are related by ratios, the ratios are real numbers, and measurement is trying to discover these real numbers for some cases. >

> eg 1 meter is a unit, & a measure of any object’s length can be expressed as some ratio of this unit (eg 4m) such that any other object will have length related to the first object also by a ratio (eg 8m which is 2x the first object’s length). >

> In contrast, in psychology ‘Measurement is the process of assigning numerals to objects or events according to rules’. In physical sciences, therefore, numbers aren’t assigned to anything but rather measurement is the *discovering* of ratios of thing’s relative to some unit.

“it is not that my room or its length is related to the number four, the length of my room relative to the metre simply is the number four”.

one explanation for psychology’s belief “that science and measurement are inextricably linked” & “that measurement is necessary” is that >

“The spectacular development of quantitative science from the seventeenth to the nineteenth centuries made it appear that the progress of science and the development of measurement went hand in hand” leading to the false premise that all science is quantitative because of” >

“the mistaken view that a method which applies to only some scientific disciplines (albeit those once thought of as paradigms of science), should characterise all scientific disciplines” >

finally culminating into the “practical syllogism: all science is quantitative; psychology aspires to be a science; therefore, psychology must be quantitative”.

But also, these views “were widely accepted within the nineteenth-century scientific community [which] meant that if psychology was to be ‘sold’ within that milieu as a science and as a science-based profession, it had to be ‘packaged’ as quantitative. >

“In this way, both scientism and practicalism motivated psychologists to adopt the measurability thesis opportunistically, as an ideological means to a social end.”

Michell says that "Numerical relations [between different magnitudes of an attribute] require additive structure". E.g., lengths are additive because two lengths (a & b) can be added to make a third (c) such that the three lengths stand in additive relation to one another (a+b=c)

now I mix it up with some of Michell's papers. It’s a mistake to think of measurement as atheoretical. The claim that any (psychological) attribute is quantitative is an empirical hypothesis embedded within “a much wider quantitative theory involving the hypothesis that >

specific quantitative relationships between attributes obtain”. This empirical claim may, in principle, be false. So, any scientist proposing such an hypothesis (i.e., that an attribute is quantitative & thus measurable) is always logically committed to the task of testing it >

In other words, when an attribute is conceptualized as quantitative, "a scientific hypothesis is proposed … [and] accepting this hypothesis is speculative, unless there is evidence specifically supporting it."

He concludes this section by alluding to what he thinks of much of psychology: “Establishing a quantitative science involves two tasks. First, … [the scientific task] of experimentally investigating the hypothesis that the relevant attribute is quantitative. >

Second, … the instrumental task of devising procedures to measure magnitudes of the attribute … [because the first task comes logically before the second] >

failure to confirm the hypothesis that the relevant attribute is quantitative [first task] means that treating the proposed measurement procedures [from the second task] as if they really are measurement procedures is at best speculation and, at worst, a pretence at science."

"psychology ... has its own definition of measurement ... quite unlike the traditional concept used in the physical sciences" which has resulted in a continuing blind spot as to "the true nature of scientific measurement" (Michell, 1997).

"As the founding father of quantitative psychology, Fechner's work established the quantitative paradigm in psychology and became the definitive exemplar emulated by others", but he ignored the first task of investigating the hypothesis that an attribute is quantitative, >

"Psychologists after Fechner also ignored the first task and concentrated upon the second, constructing number-generating procedures which, they thought, measured psychological attributes".

"[Thorndike] believed [incorrectly] that 'Measurement by relative position [as in psychology] ... gives as true, and may give as exact, a means of measurement as that by units of amount [as in physics]'...giving psychologists permission to use the term measurement for practices >

which were not supported by any scientific evidence of quantity. .. [and so] decades prior to the publication of Stevens' definition [of measurement as used in psychology], the practices of applied psychologists already conformed to it."

"there remained within psychology a critical trajectory [about measurement & whether psychological attributes were quantitative] that surfaced from time to time ... The mainstream of quantitative psychologists paid it no heed, however. As long as the criticism was internal >

".. and consisted of only a few voices, it could be ignored with impunity. However, ... [in 1940 the British Association for the Advancement of Science published it's final report with the criticism] that the additivity of psychological attributes had not been displayed >

"and, so, there was no evidence to support the hypothesis that psychophysical methods *measured* anything." After this psychologists couldn't ignore the criticisms. But instead of admitting the validity of the criticisms & setting about testing the claims that >

psychological attributes are actually quantitative & have additive structure, the combined pressures of scientism & practicalism led psychologists to frantically defend their practices & even to >

"redefine measurement in a way that legitimized psychology's claim to the scientific high ground it had never actually occupied". This resulted in a definition of measurement (in psychology) that stands apart from the definition used in physical sciences (see earlier in thread).

Because it perfectly matched what psychologists were already doing, Stevens's redefinition of measurement was quickly "cited in major texts as the only definition of measurement [&] referred to in leading journals as the 'classical' theory of measurement" -

thus were the beginnings of psychology's blind spot regarding scientific measurement -- the scientific necessity of testing via experiment that psychological attributes are quantitative.

so how do you test the claim that a psychological attribute is quantitative? Luce & Tukey's (1964) "theory of conjoint measurement ... and its subsequent developments ... revealed a range of decisive, indirect tests for the hypothesis that attributes are quantitative. ... [but] >

"it inspired only a relatively small number of empirical studies towards that end, leaving that task still seriously incomplete". - as a side note, i need to read up about conjoint measurement (no idea what it is) and hopefully understand it.