Comparability of Scores and Passing Decisions for Different Item Pools of Computerized Adaptive Examinations

• In Event: 70.066 - Roundtable Session 28
  In Roundtable Session: 70.066-7 - Equating: Issues and Practices

Abstract

Many computerized adaptive testing (CAT) programs employ multiple item pools and rotate them in use in order to address security concerns. To ensure the fairness of the assessment, proficiency estimates and passing decisions across different item pools should be psychometrically comparable. This study assesses psychometric comparability of multiple item pools for two operational CAT examinations. In addition, detailed computational procedures are provided for the statistics used for assessing such comparability. The preliminary results show that for each examination, the scores and passing decisions are psychometrically comparable across four different pools.

Authors

• Shichao Wang, University of Iowa

• Wei Wang, Educational Testing Service

• Michael J. Kolen, University of Iowa

• Won-Chan Lee, University of Iowa