A Mixture Item Response Theory Model of a Mathematics Problem-Solving Test on Speed

• In Event: Poster Session 9
  In Poster Session: New Developments in Psychometrics, Measurement, and Assessment: Poster Session 1

Abstract

Word problems on speed were frequently included in international studies like TIMSS and PISA, indicating the significance of the speed topic in the school curriculum. This study used the mixture IRT model to determine whether there exists two or more distinct sub-populations with respect to the measurement of math word problems on speed. Two distinct classes emerged. The two latent classes of students may employ different strategies for solving math word problems which could affect their performance. The mixed IRT model provided more insight about cross-culture differences in mathematics achievement stem from curricular differences. The findings have clear instruction implications for teachers in planning and individualizing instruction. They also offer insights and implications for designing and conducting cross-culture research.

Authors

• Do-Hong Kim, University of North Carolina - Charlotte

• Chuang Wang, University of North Carolina - Charlotte

• Chunlian Jiang, University of Macau