Exploring the Impact of Missing Data on Principal Component Analysis of Residuals

• In Event: Methodological Approaches in Rasch Measurement

Abstract

Researchers frequently use Rasch models to analyze survey responses because these models provide accurate parameter estimates for items and examinees when there is missing data. However, researchers have not considered how missing data affect the accuracy of dimensionality assessment tools for Rasch analyses such as principal components analysis (PCA) of standardized residuals. Because adherence to unidimensionality is a prerequisite for the appropriate interpretation and use of Rasch model results, insight into the impact of missing data on the accuracy of this approach is critical. We used a simulation study to examine the accuracy of PCA of standardized residuals with various missing data proportions and multidimensionality. Our results suggested that missing data impacts the accuracy of PCA of standardized residuals.

Authors

• Stefanie A. Wind, The University of Alabama - Tuscaloosa

• Randall E. Schumacker, The University of Alabama