A generalized dimensionality discrepancy measure is introduced to facilitate a critique of dimensionality assumptions in multidimensional item response models. Connections between dimensionality and local independence motivate the development of the discrepancy measure from a conditional covariance theory perspective. A simulation study and a real-data analysis demonstrate the utility of the discrepancy measure's application at multiple levels of analysis in a posterior predictive model checking framework.
|Number of pages
|British Journal of Mathematical and Statistical Psychology
|Published - May 2011
ASJC Scopus subject areas
- Statistics and Probability
- Arts and Humanities (miscellaneous)