A Multidimensional Model to Facilitate Within Person Comparison of Attributes

Area of Science:

• Psychometrics
• Psychological Measurement
• Conjoint Measurement

Background:

• Within-person comparisons in psychology often assume equal units (EU) or origins, an assumption rarely validated.
• Existing measurement models may not guarantee the necessary properties for valid score comparisons.

Purpose of the Study:

• To describe a multidimensional SEM/IRT model applicable to conjoint measurement principles.
• To demonstrate how this model can establish equal units (EU) for psychological measurements.
• To enhance the validity of psychological scores for within-person comparisons.

Main Methods:

• Utilized a multidimensional Structural Equation Modeling/Item Response Theory (SEM/IRT) model.
• Applied principles of conjoint measurement to validate the model's expected response variables.
• Employed Equal Units (EU) analysis as a pre-processing step for Quality of Life data.

Main Results:

• The described SEM/IRT model, via conjoint measurement, ensures response variables satisfy axioms for measurement on a common scale.
• EU analysis successfully derived a simple structure Quality of Life model with three dimensions in equal units.
• Validated EU analysis demonstrated added validity for scores, enabling new research questions.

Conclusions:

• The proposed SEM/IRT model and EU analysis provide a rigorous framework for establishing equal units in psychological measurement.
• When the EU model fits data, derived scores possess enhanced validity for addressing complex within-person comparisons.
• Further research is needed to explore limitations and expand the application of this methodology.