Measuring the Unmeasurable: The Psychometrics of Life History Strategy

S.L.K. Gruijters, Bram P.I. Fleuren

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Within evolutionary biology, life-history theory is used to explain cross-species differences in allocation strategies regarding reproduction, maturation, and survival. Behavioral scientists have recently begun to conceptualize such strategies as a within-species individual characteristic that is predictive of behavior. Although life history theory provides an important framework for behavioral scientists, the psychometric approach to life-history strategy measurement—as operationalized by K-factors—involves conceptual entanglements. We argue that current psychometric approaches attempting to identify K-factors are based on an unwarranted conflation of functional descriptions and proximate mechanisms—a conceptual mix-up that may generate unviable hypotheses and invites misinterpretation of empirical findings. The assumptions underlying generic psychometric methodology do not allow measurement of functionally defined variables; rather these methods are confined to Mayr’s proximate causal realm. We therefore conclude that K-factor scales lack validity, and that life history strategy cannot be identified with psychometrics as usual. To align theory with methodology, suggestions for alternative methods and new avenues are proposed.
Original languageEnglish
Pages (from-to)33-44
Number of pages12
JournalHuman Nature-An Interdisciplinary Biosocial Perspective
Volume29
Issue number1
DOIs
Publication statusPublished - Mar 2018

    Fingerprint

Keywords

  • Life history strategy
  • Ultimate-proximate distinction
  • Measurement models
  • Psychometrics
  • Formative models
  • Latent variables
  • Validity
  • MEASUREMENT MODELS
  • K-FACTOR
  • MULTIDIMENSIONAL CONSTRUCTS
  • INDIVIDUAL-DIFFERENCES
  • REPRODUCTIVE STRATEGY
  • FORMATIVE MEASUREMENT
  • EVOLUTIONARY-THEORY
  • CAUSAL INDICATORS
  • LATENT-VARIABLES
  • TRADE-OFFS

Cite this