Skip navigation

Psychometrics is not measurement: Unraveling a fundamental misconception in quantitative psychology and the complex network of its underlying fallacies

Psychometrics is not measurement: Unraveling a fundamental misconception in quantitative psychology and the complex network of its underlying fallacies

Uher, Jana ORCID: 0000-0003-2450-4943 (2020) Psychometrics is not measurement: Unraveling a fundamental misconception in quantitative psychology and the complex network of its underlying fallacies. Journal of Theoretical and Philosophical Psychology, 41 (1). pp. 58-84. ISSN 1068-8471 (Print), 2151-3341 (Online) (doi:https://doi.org/10.1037/teo0000176)

[img]
Preview
PDF (Author Accepted Manuscript)
30461 UHER_Psychometrics_is_Not_Measurement_(AAM)_2020.pdf - Accepted Version

Download (1MB) | Preview

Abstract

Psychometrics has always been confronted with fundamental criticism, highlighting serious insufficiencies and fallacies. Many fallacies persist, however, because each critic explores only some fallacies while still building on others. This article scrutinizes the epistemological, metatheoretical, and methodological foundations of psychometrics, revealing a complex network of numerous conceptual fallacies underlying its framework of theory and practice. At its core lies a key challenge for psychology: the necessity to distinguish the phenomena under study from the means used to explore them (e.g., concepts, methods, data). This distinction is intricate because concepts constitute psychical phenomena in themselves and many psychical phenomena are accessible only through language-based methods. The analyses show how insufficient consideration of this important distinction and common misconceptions about concepts and language (e.g., signifier–referent conflation, reification of constructs) led to conflations of disparate notions of key terms in psychological measurement (e.g., “variables”, “attributes”, “causality”) and numerous interrelated fallacies (e.g., construct–referent conflation, phenomenon–quality–quantity conflation, numeral–number conflation). These fallacies are maintained and masked by repeated conceptual back-and-forth switching between two incompatible epistemological frameworks, (a) an operationist framework of data modeling implemented through methodical and statistical operations and (b) a realist framework of measurement sporadically invoked in theoretical considerations but neither theoretically elaborated nor empirically implemented. The analyses demonstrate that psychometrics constitutes only data modeling but not data generation or even measurement as often assumed and that analogies to (indirect or fundamental) physical measurement are mistaken. They provide theoretical support for the increasing criticism of psychometrics and its use in research and applied contexts.

Impact Statement
Public Significance Statement—The idea of psychometrics as enabling science-based quantifications of the mind is shown to be based on ambiguous meanings of key terms that are often confused with one another as well as on erroneous assumptions about how measurement can be implemented in research about the mind. The analyses show that psychometrics does not establish systematic relations to individuals’ minds as needed for measurement and that, consequently, psychometric results should not be used to make decisions about persons.

Item Type: Article
Uncontrolled Keywords: psychometrics, replicability, latent variable, psychological measurement, quantitative method
Subjects: B Philosophy. Psychology. Religion > BF Psychology
Faculty / Department / Research Group: Faculty of Education, Health & Human Sciences
Faculty of Education, Health & Human Sciences > Department of Psychology, Social Work & Counselling
Related URLs:
Last Modified: 22 Mar 2021 21:10
Selected for GREAT 2016: None
Selected for GREAT 2017: None
Selected for GREAT 2018: None
Selected for GREAT 2019: None
Selected for REF2021: None
URI: http://gala.gre.ac.uk/id/eprint/30461

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics