Improving the Measurement of Students’ Composite Ability Score in Mixed-Format Assessments

• DOI: 10.3390/educsci15030374
• Publication Date: 2025-03-18T14:43:51Z
• Authors: Jiawei Xiong; Qidi Liu; Cheng Tang; Bowen Wang; Allan S. Cohen

Mixed-format assessments, which include both multiple-choice (MC) and constructed-response (CR) items, often produce separate scoring scales, with MC items scored dichotomously CR polytomously. Conventional methods for estimating composite ability scores, such as weighting or summing, rely on subject matter expertise but overlook the information embedded in item scores. While recent progress takes advantage of empirical Bayes analysis it may also introduce biases because relies solely point estimates without accounting variability unknown distributions other parameters. To address these gaps, this study introduces a practical easily implementable method, fully Bayesian, that leverages scores to derive priors, leading more accurate score estimates. It has been found have effectively captured students’ achievement assessment domain, they could provide valuable final scoring. Through analyses students Grades 3 10 two additional simulation studies based real-world data, we demonstrate approach enhances reliability, reduces reporting biases, provides evaluation tool mixed-format assessments.