Psychometrics is one of the main approaches to social scientific measurement. It is relied upon in drug testing, economic policymaking, recruitment, and other decision-making contexts. The first aim of this article is to introduce philosophers to key aspects of psychometrics, namely, classical test theory, item response theory, and construct validity. The second aim is to show how a debate on the nature of psychological attributes manifests in psychometrics. In this debate, realists claim that psychometric measures are indicators of independently existing (...) qualities, while operationalists argue that a measure defines its target attribute. The third and final aim is to argue that despite its poor reputation among philosophers, operationalism is a viable approach to psychometrics. (shrink)

Are widely used measurements in the human sciences (say happiness surveys or depression scales) quantitative or merely ordinal? If they are merely ordinal, could they be developed into quantitative measurements, just like in the progression from thermoscopes to thermometers? Taking inspiration from recent philosophy of measurement, some practitioners express optimism about future human science measurements. The so-called quantity objection stands out for having the only chance of settling the debate in favour of the pessimists. It claims that the problem lies (...) not with current, or likely future, measurement practices in the human sciences, but with human science attributes themselves—they just are not quantitative, but merely ordinal. Hence, they cannot not (thus will not) be measured quantitatively. The argument has a long and distinguished pedigree. This paper assesses old and recent versions of it, namely: the objection made originally to Fechner’s psychophysics by von Kries (among others) and Michell’s recent version of this objection. To do so, the paper first draws important distinctions between different versions of the argument that have been overlooked. Then, it argues that none of the versions of the quantity objection provide a good reason for the optimists to give up their optimism. In particular, Michell’s argument characterizes the measurand (that is, the attribute to be measured) in a way that optimists do not and need not accept. Yet the optimists’ defence articulated here brings with it serious burdens to discharge. The ball is in their court. (shrink)

The article advances a new way of thinking about classifications in general and the classification of mental disorders in particular. By applying insights from measurement practice to the context of classification, I defend a notion of epistemic accuracy that allows one to evaluate and improve classifications by comparing different classifying methods to each other. Progress in classification arises from the mutual development of classification systems and classifying methods. Based on this notion of accuracy, the article illustrates with an example how (...) psychiatric classifications can be improved via circumscribed comparisons of different perspectives on mental disorders, without relying on complete models of their complex aetiology. When applying this strategy, the traditional opposition between symptom-based and causal approaches is of little consequence for making progress in the epistemic accuracy of psychiatric classification. (shrink)

In this dissertation, I present a novel account of the components that have a peculiar epistemic role in our scientific inquiries, since they contribute to establishing a form of coordination. The issue of coordination is a classic epistemic problem concerning how we justify our use of abstract conceptual tools to represent concrete phenomena. For instance, how could we get to represent universal gravitation as a mathematical formula or temperature by means of a numerical scale? This problem is particularly pressing when (...) justification for using these abstract tools comes, in part or entirely, from knowledge which is not independent from them, thus leading to threats of circularity. Achieving coordination between some abstract conceptual tools and the concrete phenomena that they are supposed to represent is usually a complex process, which involves several epistemic components. Some of these components eventually provide stable conditions for applying those abstract representations to concrete phenomena. It is in this sense of providing certain conditions of applicability that different philosophical traditions, as well as some contemporary reappraisals, view these components as constitutive or a priori. In this work, I present a new gradualist, contextualist, and relational approach to understand these constitutive components of scientific inquiry. It is gradualist inasmuch as the degree to which some component is constitutive depends on three quantifiable features: quasi-axiomaticity, generative potential, and empirical shielding. Since the quantification of these three features impinges on the history and practice of using these components in a scientific context, my approach is a contextualist one. Finally, my approach is relational in a double sense: first, it identifies ordinal relationships among epistemic components with respect to their constitutive character; second, these relationships are relative to a scientific framework of inquiry. After introducing my account and a classic example of constitutively a priori principles, i.e., Friedman’s (2001) analysis of Newtonian mechanics, I turn to my own case studies to demonstrate the advantages of my approach. Firstly, I discuss Okasha’s (2018) view of endogenization as a pervasive theoretical strategy in evolutionary biology and suggest that the constitutive character of the core Darwinian principles progressively increases with endogenization. Secondly, I apply a conceptual distinction between two varieties or scopes of coordination – general coordination and coordination in measurement – to Ohm’s work on electrical conductivity. This distinction allows me to pinpoint to what extent components along different dimensions (e.g., instrumentation, measurement, theorising, etc.) were constitutive of the forms of coordination which Ohm relied on. Thirdly, I discuss the epistemic function of the Hardy-Weinberg principle in the history and practice of population genetics. I assess this principle in terms of my account and identify approximation and stability as two components that are highly constitutive, in that they contribute to justifying its use in population genetics. Finally, applying my account to these case studies enables me to identify at least three qualitatively different types of constitutive components: domain-specific theoretical principles, material components, and domain-independent assumptions underlying reasoning abilities. In the light of my results, I draw some general conclusions on epistemic justification and scientific knowledge. (shrink)

Since the 1970’s epidemiological measures focusing on “health-related quality of life” or simply “quality of life” have figured increasingly as endpoints in clinical trials. Before the 1970’s these measures were known, generically, as performance measures or health status measures. Relabeled as “quality of life measures” they were first used in cancer trials. In the early 2000’s they were relabeled again as “patient-reported outcome measures” or PROMs, in their service to the FDA to support drug labeling claims. To the limited degree (...) that the philosophical literature addresses these measures, it tends to associate them with two major theories of well-being: subjective well-being and the capabilities approach. In this paper, I argue that philosophers ought to treat quality of life measures/proms as a distinct from these theories. I argue that quality of life/proms serve healthcare as vehicles for patient-centered care and this has important consequences for how we theorize these measures. To this end, two requirements set them apart from measures of subjective well-being and the capabilities approach: quality of life/proms need to be patient-directed and inclusive. (shrink)

Psychometrics and the representational theory of measurement are widely used in social scientific measurement. They are currently pursued largely in isolation from one another. I argue that despite their separation in practice, RTM and psychometrics are complementary approaches, because they can contribute in complementary ways to the establishment of what I argue is a crucial measurement property, namely, representational interpretability. Because RTM and psychometrics are complementary in the establishment of representational interpretability, the current separation of measurement approaches is unfounded. 1Introduction2Two (...) Approaches to Measurement 2.1Representational theory of measurement2.2Psychometrics2.3Representational interpretability3Complementarity, Conceptually 3.1Representational theory of measurement: Conditions of representational interpretability3.2Psychometrics: Evidence of representational interpretability4Complementarity in Action 4.1What is the Rasch model?4.2Rasch and conjoint measurement5Conclusion: Critics and Fruits of Complementarity. (shrink)

The Representational Theory of Measurement is the best known account of the kind of representation measurement requires. However, RTM has been challenged from various angles, with critics claiming e.g. that RTM fails to account for actual measurement practice and that it is ambiguous about the nature of measurable attributes. In this paper I use the critical literature on RTM to formulate Representation Minimalism – a characterization of what measurement-relevant representation requires at the minimum. I argue that Representation Minimalism avoids the (...) main problems with RTM while acknowledging its usefulness as the formal foundation of representation in measurement. (shrink)

This paper applies conceptual engineering to deal with four objections that have been levelled against operationalism in psychology. These objections are: operationalism leads to harmful proliferation of concepts, operationalism goes hand-in-hand with untenable antirealism, operationalism leads to arbitrariness in scientific concept formation, and operationalism is incompatible with the usual conception of scientific measurement. Relying on a formulation of three principles of conceptual engineering, I will argue that there is a useful form of operationalism that does not fall prey to these (...) four objections. (shrink)