Methodological empiricism and the choice of measurement models in social sciences

Abstract

Realism is generally assumed as the correct position with regards to psychological research and the measurement of psychological attributes in psychometrics. Borsboom et al. (2003), for instance, argued that the choice of a reflective measurement model necessarily implies a commitment to the existence of psychological constructs as well as a commitment to the belief that empirical testing of measurement models can justify their correspondence with real causal structures. Hood (2013) deemphasized Borsboom et al.'s position and argued that the choice of a reflective measurement model does not necessarily require ontological commitments, though, in his view, it does necessitate a commitment to minimal epistemic realism. Although these arguments are formulated with regard to psychological research, they can actually be generalized to other disciplines in social sciences that use similar methodologies and statistical techniques. In Hood's opinion, empiricism does not suffice to provide an adequate account of the choice of reflective measurement models given that this choice requires an appeal to causal explanations. In this paper, we argue against Hood and answer this challenge, providing epistemic foundations for social science research that do not appeal to realism.

Operationalism and realism in psychometrics

“…2; Meehl, ) . Accordingly, Peterson () writes that “[r]ealism is generally assumed as the correct position with regards to psychological research and the measurement of psychological attributes in psychometrics.” Nonetheless, operationalism seems to permeate the practice of psychological measurement, that is, operationalist inclinations are implicit in the way psychometric measures are constructed and tested. Michell (, ), for example, argues that Stevens' operationalism lives on in contemporary practices of measure validation.…”

Respectful operationalism

“…How should we handle the plurality of related concepts that likely ensues when we allow respectfully operationalist and realist validation to co-exist? To answer these and other open questions, I believe we need a combination of literature on concept formation and measure validation (e.g., Alexandrova & Haybron, 2016; Angner, 2013; McClimans, 2013; Peterson, 2018), literature on psychometric validation (e.g., Borsboom, 2005; Cronbach & Meehl, 1955; Fiske, 1971; Loevinger, 1947), and studies of the history of operationalism (e.g., Chang, 2017b; Feest, 2005; Rogers, 1989).…”