Bundle Models for Computerized Adaptive Testing in E-Learning Assessment
Kathleen Scalise · 2007
A multifacet bundle model, herein called the “iota model ” was used to estimate "pathway" parameters through partially hierarchical testlets. The model is useful for computerized adaptive assessment in e-learning contexts, when students are receiving individualized, or personalized, delivery of content based on embedded assessments. Testlets in this case are small bundles of items that act as questions and follow-up probes to interactively measure and assign scores to students. Research considered whether testlets can serve as a valid and reliable design to collect data and implement interactions for personalized delivery of content, whether path scores through the testlet modeled to a cognitive framework can be considered equivalent, and how three testlet designs compared in the quality and consistency of data collected. An example is shown that is multistage CAT, in which sequential or preplanned pathways are adaptively presented to students within the testlets based on student responses, and updating of θ and standard CAT algorithms can be used between the testlets.