Extracting Students’ Self-Regulation Strategies in an Online Extensive Reading Environment using the Experience API (xAPI)

Abstract

Extensive Reading (ER) activity is useful in language learning where learners pick any reading materials in target language by themselves and continue reading. In this study, we aim to understand students’ self-regulation behaviors and strategies in ER so that facilitators can give learning strategy-based instruction. For this purpose, we first explored a potential structure of a learners’ model that can highlight their learning strategies and extracted from multi-system interaction logs recorded in Experience API (xAPI) format. To demonstrate, we collected data from 120 students across 3 months when they did ER activities. We analyzed and extracted the self-regulation strategies of the learners and found 2 groups by applying K-means cluster analysis. The results inform dashboard design and instructional support based on the visualized attributes of the cluster members. This study contributes towards using interoperability standards to record learner’s online reading behaviors and demonstrate how teaching and learning activities can be supported by xAPI when such experiences are distributed across various learning tools.