Date of Award

Fall 11-2023

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Information Systems (PhDIS)

Department

Business and Information Systems

First Advisor

Cherie Noteboom

Second Advisor

David Zeng

Third Advisor

David Zeng

Abstract

Recommender systems, or recommenders, are information filtering systems prevalent today in many fields. One type of recommender found in the field of education, the educational recommender, is a key component of adaptive learning solutions as these systems avoid “one-size-fits-all” approaches by tailoring the learning process to the needs of individual learners. To function, these systems utilize learning analytics in a student-facing manner.

While existing research has shown promise and explores a variety of types of educational recommenders, there is currently a lack of research that ties educational theory to the design and implementation of these systems. The theory considered here, self-regulated learning, is underexplored in educational recommender research. Self-regulated learning advocates a cyclical feedback loop that focuses on putting students in control of their learning with consideration for activities such as goal setting, selection of learning strategies, and monitoring of one’s performance.

The goal of this research is to explore how best to build a self-regulated learning guided educational recommender and discover its influence on academic success. This research applies a design science methodology in the creation of a novel educational recommender framework with a theoretical base in self-regulated learning. Guided by existing research, it advocates for a hybrid recommender approach consisting of knowledge-based and collaborative filtering, made possible by supporting ontologies that represent the learner, learning objects, and learner actions. This research also incorporates existing Information Systems (IS) theory in the evaluation, drawing further connections between these systems and the field of IS. The self-regulated learning-based recommender framework is evaluated in a higher education environment via a web-based demonstration in several case study instances using mixed-method analysis to determine this approach’s fit and perceived impact on academic success. Results indicate that the self-regulated learning-based approach demonstrated a technology fit that was positively related to student academic performance while student comments illuminated many advantages to this approach, such as its ability to focus and support various studying efforts. In addition to

contributing to the field of IS research by delivering an innovative framework and demonstration, this research also results in self-regulated learning-based educational recommender design principles that serve to guide both future researchers and practitioners in IS

and education.

Download

Share

COinS