Item-Explanatory Mokken Scaling: Using Nonparametric Item Response Theory to Explore Item Characteristics

• In Event: Applications in Psychometric Models

Abstract

Item-explanatory measurement models are an active area in methodological and applied measurement research, relying exclusively on parametric item response theory (IRT) models such as the Many-Facet Rasch model and explanatory IRT. We considered how nonparametric IRT techniques based on Mokken Scale Analysis (MSA) can be used to explore the effects of item characteristics with an emphasis on invariance. We used two real-data illustrations and evaluated the sensitivity of the approach using a simulation. Results suggested that MSA techniques can be extended to examine item properties for evidence of monotonicity and invariant ordering. This technique provides an exploratory tool that complements and supplements standard nonparametric IRT procedures for evaluating item properties.

Author

• Stefanie A. Wind, University of Alabama