Concept-Based Learning in Blended Environments Using Intelligent Tutoring Systems

摘要

CoLaB Tutor and AC-ware Tutor are Intelligent Tutoring Systems (ITSs) that are based on concept-based learning and are notable due to the fact they are relatively easy to generalize to multiple knowledge domains. In this research study we investigate the performance of CoLaB Tutor, AC-ware Tutor, and Moodle in a blended learning environment for an introductory computer programming course. In our study, regular face-to-face lectures and laboratory exercises were complemented with online learning at the students' own pace, time and location. Our study revealed that CoLaB Tutor students had moderately higher knowledge gains than those students in the AC-ware and Moodle groups. The prediction of student success (pass/fail) for a basic knowledge post-test revealed an overall classification rate of 73,5% for the CoLaB Tutor group (completed knowledge and online score as predictors), 71,4% for the AC-ware group (completed knowledge as predictor) and 70% for the Moodle group (time spent online as predictor). Additionally, students that used ITSs on average passed through more knowledge online than students that used LMS, while students that used LMS on average spent more time online.