A Meta-Analysis of Adaptive Learning Systems on Learning Achievement (Poster 16)

• In Event: Friday Poster Session 9:35 am
  In Poster Session: Trends and Issues in Technology-Based Environments

Abstract

Adaptive learning that provides customized learning tasks based on learners’ characteristics (e.g., learning style, prior knowledge) has been an innovative and promising area in education. Research has suggested varying effectiveness of adaptive learning systems (ALSs) on learning achievement. This meta-analysis included 46 studies to synthesize the effectiveness of ALSs on student achievement and revealed a robust and positive relationship between the use of ALSs and learning achievement improvement, g = 1.227 (CI: 0.832–1.622). We examined moderator effects of learner characteristics and modeling approaches: ALSs incorporating both learning style and prior knowledge yielded larger learning gains than those considering one of the characteristics. Learning gains were more pronounced when modeling prior knowledge using educational data mining compared to other techniques.

Authors

• Yizhu Gao, University of Georgia

• Ying Cui, University of Alberta

• Xiaoming Zhai, University of Georgia