Improving Long-Term Retention through Personalized Recall Testing and Immediate Feedback

Tests have been shown to improve the long-term retention of students, referred to as the "testing effect". Different question formats, such as multiple choice and recall, have varying properties. While recall questions induce a longer retention than multiple choice ones, they are harder for students since answers must be actively retrieved from memory. Moreover, regular testing inherently increases the teacher's workload, particularly since feedback on each test is paramount for an effective bearing on the student's understanding and learning. This work introduces BEACON Q, a digital quiz application combining different question formats and progressively adapting their difficulty to each student’s level. The level is derived from past answers as well as ratings provided by the students which constitute their perceived understanding of each topic. BEACON Q delivers immediate feedback through detailed explanations, without requiring manual assessment by the teacher. Tests are scheduled for periods of time, thus giving the students the flexibility to choose an appropriate time to take the test. An initial evaluation of BEACON Q has been performed in the context of three different computer science classes at our university. Preliminary results are presented in this paper.