Using the Mixture Rasch Model to Explore Knowledge Resources Students Invoke in Mathematic and Science Assessments

• In Event: Poster Session 9
  In Poster Session: Applications of Cognitive Diagnostic Models: From Q-Matrix Development to Data-Model Fit

Abstract

The purpose of the study was to investigate whether mixture Rasch models followed by qualitative item-by-item analysis of PISA (2009) selected mathematics and science items offered insight into knowledge students invoke in mathematics and science separately, and combined. The researchers administered an assessment constructed from PISA released items to 516 students in China. The findings suggest that while PISA attributes showed promise for providing insight into how students were classified in mathematics and science, when combined these attributes were not found useful. Our findings suggest that students do not seem to be applying attribute strengths to the dataset as a whole (i.e., mathematics and science items combined) in ways that differentiate them from students who appear weaker for those attributes.

Authors

• Danhui Zhang, Beijing Normal University

• Chandra H. Orrill, University of Massachusetts - Dartmouth

• Todd Campbell, University of Connecticut