Comparing Pretest Item Selection Methods in Online Calibration for Cognitive Diagnostic Models

• In Event: 14.046 - Poster Session 1
  In Poster Session: 14.046-6 - Contemporary Models and Their Applications: A Poster Session

Abstract

Item parameter calibration is important in item response theory based tests, because procedures including scoring, equating, DIF analysis, and item selection in adaptive tests are all based on value of item parameters. Among various calibration approaches, online calibration dynamically selects the pretest items for each examinee during the operational test, which could improve the efficiency of calibration. This study compares different pretest item selection methods used for online calibration under a computerized adaptive testing design. Results show that item-centered selection methods outperformed examinee-centered methods.

Authors

• Rui Guo, University of Illinois at Urbana-Champaign

• Hua-Hua Chang, University of Illinois at Urbana-Champaign