Improved Measures with the Experimental Psychometrics Framework

This paper proposes the experimental psychometrics framework, an approach that integrates causal inference and experimental design, representational measurement theory, and data theory to enhance psychological measurement. The framework addresses limitations of traditional psychometric methods by emphasizing the need for experimental control to account for sources of systematic error and to test the assumptions underlying psychometric models. It is argued that past propositions found in the literature have demonstrated how experimental manipulation can improve the validity of measures and provide better tests of theoretical expectations. However, the application of experimental manipulation and causal modeling as tool for assessing the abstract structure of the data, that can be tested from a representational measurement theory perspective are not well developed. Two empirical examples are presented to illustrate how the experimental psychometrics framework offers a structured approach for deriving more robust and replicable measurements. This work contributes to the growing discourse on improving the rigor of psychological measurement and theory, advocating for a more experimentally-driven and theory-informed approach. The potential for this framework to reshape psychometric practices is significant, promising to enhance the accuracy, validity, and reliability of psychological assessments.