Point-to-Point: A Reply to “A Critique to ‘Using Rasch Analysis to Create and Evaluate a Measurement Instrument for Foreign Language Classroom Speaking Anxiety’ ”

Abstract

In his response to my paper concerning the use of the Rasch model for creating and evaluating foreign language classroom speaking anxiety (Apple, 2013), Dr. Panayides makes some interesting observations; however, there also appear to be several points of misinterpretation of the study results. The initial issue is his opening assertion that my paper was designed to show advantages of the Rasch model over classical test theory (CTT) models as well as item response theory (IRT) models. In fact, the paper was designed only to demonstrate the advantages of the Rasch model for Japan-based classroom teachers of English. I made no mention whatsoever of other IRT models. I also did not set the Rasch model against all CTT methods; I merely demonstrated that simple descriptive statistics were not as informative or useful as Rasch analysis when creating and evaluating questionnaires.