Adaptive Testing for Multiple-Choice, Multiple-Attempt Test Items

• In Event: Computer Adaptive Testing Applications

Abstract

This study proposes the combination of multiple-choice, multiple-attempt test items and CAT (MM-CAT). Our small-scale simulation study demonstrated that CAT can benefit from allowing multiple attempts on items by either improving the accuracy of theta estimates in fixed-length CAT or reducing the test length of variable-length CAT.

Authors

• Yikai Lu, UNIVERSITY OF NOTRE DAME

• Ying Cheng, UNIVERSITY OF NOTRE DAME