The Reliability of Diagnosing Broad and Narrow Skills in Middle School Mathematics with the Multicomponent Latent Trait Model
The multicomponent latent trait model for diagnosis (MLTM-D;Embretson and Yang, Psychometrika 78:14–36, 2013) is a conjunctive item response model that is hierarchically organized to include broad and narrow skills. A two-stage adaptive testing procedure was applied to diagnose skill mastery in middle school mathematics and then analyzed with MLTM-D. Strong support for the reliability of diagnosing both broad and narrow skills was obtained from both stages of testing using decision confidence indices.
KeywordsDiagnostic models Item response theory Multidimensional models Decision confidence reliability
- Du Toit, M. (2003). IRT from SSI. Lincolnwood, IL: Scientific Software International, Inc.Google Scholar
- Embretson, S. E. (2015). The multicomponent latent trait model for diagnosis: Applications to heterogeneous test domains. Invited paper for Applied Psychological Measurement (pp. 6–30).Google Scholar
- Embretson, S. E., Morrison, K., & Jun, H. (2014). Implementation of a multistage adaptive testing system for diagnosing mathematical deficits in middle school. Report IES1001A-2014 for Institute of Educational Sciences Grant R305A100234. Atlanta, GA: Cognitive Measurement Laboratory, Georgia Institute of Technology.Google Scholar
- Rudner, L. M. (2005). Expected classification accuracy. Practical Assessment Research & Evaluation, 10, 1–4.Google Scholar
- Rupp, A. A., Templin, J., & Hensen, R. A. (2010). Diagnostic measurement: Theory, methods, and applications. New York: Guilford Press.Google Scholar