Identifying Misfitting Achievement Estimates in Performance Assessments: An Illustration Using Rasch and Mokken Scale Analyses

• In Event: 56.082 - Poster Session 12
  In Poster Session: 56.082-3 - Current Issues in School-Based Assessment

Abstract

Person fit analyses are used to explore the degree to which a person’s test score is a reasonable estimate of their achievement level. Recent guidelines promote person fit analysis as a step to ensure data quality. In this study, an approach to examining person fit in educational assessment is illustrated. Performance assessment data (N=12,887) were analyzed with Rasch measurement theory and Mokken scale analysis. Results suggest that some persons provide responses that do not fit the assumptions of the Rasch model. For some of these persons, their responses adequately fit the less restrictive assumptions of the Mokken model. For extremely misfitting persons, neither the Rasch, nor the Mokken assumptions are met. This evidence informs score interpretation and use for individuals.

Authors

• Angela Adrienne Walker, Emory University

• Stefanie Anne Wind, The University of Alabama - Tuscaloosa