What CDM Can Tell About What Students Have Learned: An Analysis of TIMSS Eighth Grade Mathematics

Abstract

International trended assessments have long attempted to provide instructional information to educational researchers and classroom teachers. Studies have shown that traditional methods of item analysis have not provided specific information that can be directly applicable to improve student performance. To this end, cognitive diagnosis models (CDMs) were developed to identify the presence or the absence of multiple fine-grained skills or attributes. This study reanalyzed the TIMSS 2003 8th grade mathematics assessment by applying a cognitive diagnosis modeling approach to provide useful information about test takers. Specifically, the deterministic, inputs, noisy, “and” gate (DINA; Junker & Sijtsma, 2001) model was used to provide more valuable diagnostic information about how well students perform on underlying skills and cognitive processes necessary to answer mathematics problems. Results showed that the interpretable benefits of using the DINA discrimination index provided additional information about comparative differences between the U.S. and Korea.