A Note on the Use of Categorical Subscores

Although there exists an extensive amount of research on subscores and their properties, limited research has been conducted on categorical subscores and their interpretations. In this paper, we focus on the claim of Feinberg and von Davier that categorical subscores are useful for remediation and instructional purposes. We investigate this claim by examining (a) the agreement between true and observed subscore classifications and (b) the agreement between subscore classifications across parallel forms of a test. Results show that the categorical subscores of Feinberg and von Davier are often inaccurate and/or inconsistent, pointing to a lack of justification for using them for remediation or instructional purposes.