Conference Objects and Reports

A Hierarchical Mixed Membership Model for Multiple-Answer Multiple-Choice Items with Signal Detection Theory

Chen, Yi; Chae, Hui Soo; Natriello, Gary J.

Multiple-answers multiple-choice (MAMC) items are widely used in educational testing and social research. However, inadequate studies and resources are allocated to the assessments of MAMC items. In this study, we introduce a new approach to analyze MAMC items using original response data (which alternatives have been selected) without scoring (correct or wrong). It has immense potential to provide rich information relevant for tractable psychological behavior and interpretable educational measurement. We use the signal detection theory (SDT) to measure the decision-making behavior across alternatives. Then, the mixed membership model is applied to capture the grouped data structure in the MAMC item. A simulation study of the HMM-SDT model is presented with a comparison to the tradition treatments in Item Response Theory (IRT).

Files

  • thumnail for A_Hierarchical_Mixed_Membership_Model_for_Multiple_Answer_Multiple_Choice_Items_with_Signal_Detection_Theory.pdf A_Hierarchical_Mixed_Membership_Model_for_Multiple_Answer_Multiple_Choice_Items_with_Signal_Detection_Theory.pdf application/pdf 375 KB Download File

More About This Work

Academic Units
Gottesman Libraries
Human Development
Published Here
January 10, 2020

Notes

Northeastern Educational Research Association, Trumbull, CT, October 2019