Application of the Graded Response Model of Item Response Theory to Computerized Dynamic Assessment in L2 English Education

Abstract

The purposes of this paper are to (a) develop an adaptive system with the help of the graded response model of IRT, (b) test the system with a small group of L2 learners, (c) examine the feasibility of computerized dynamic assessment, and (d) propose a new set of scoring mechanisms. We developed a computer-based adaptive system that provides students with appropriate test items and a set of graduated prompts from implicit to explicit under the conditions of mediation. We found that the set of graduated prompts successfully guided students to the correct answer and provided rich diagnostic information needed for L2 English education. With the help of the graded response model of IRT, computerized dynamic assessment will become a more advanced and practical educational assessment tool in the field of L2 studies.

Constructing an Ecological System for College English Education Based on Artificial Intelligence and Environmental Responsibility

“…They also identified key markers that could be used for data analysis. Sumi and Miyazawa (2017) suggested that effective scaffolding for English language learners involves predicting the level of support needed to complete tasks and determining the appropriate level of help to facilitate learners' development. X. proposed a hybrid design of English teaching methodology courses guided by constructivist theory, and reported the corresponding norms and learner performance outcomes.…”