Towards Educational Recommender Systems for Self-Directed Learning: Profiling Student Achievement Goals and Identity Formation to Inform Recommendation Evaluation Metrics

Abstract

Research into recommender systems for educational contexts has come a long way but more advances are necessary for the application of evaluation metrics to academic choices in student-centered contexts. This qualitative study looks at a context where students make decisions for a self-directed learning project and analyzes their dispositional, contextual, and situational experiences from the theoretical perspectives of achievement goal theory and identity status theory. We interviewed 7 students at the conclusion of their projects in a semi-structured format with interview questions based on the theoretical aspects of achievement goal theory and identity status theory. Our analysis of the student responses then identifies their needs during the decision-making process and discusses the implications for effective recommender system metrics. The discussion is shaped by the general aim of designing an educational recommender system that promotes exploratory behavior in students and our analysis reveals potential user types for the design of such a system along with relevant evaluation metrics.