Using a Variety of Interactive Learning Methods to Improve Learning Effectiveness: Insights from AI Models Based on Teaching Surveys

Abstract

The last decade has brought far-reaching changes in higher education, leading institutions to shift some or all instruction online. This shift to distance learning has contributed to a more significant need for active learning: changing students from passive knowledge consumers into proactive knowledge producers using interactive teaching practices. The present study joins an emerging body of literature examining the relationship between active learning, the online environment, and students’ performance. In this study, we examined the effect of four interactive learning methods (combined with technology) on students’ overall assessments of the class, the clarity of the teaching, and the perceived effectiveness of online distance learning. The data source for the research is teaching evaluation surveys filled out by undergraduate and master’s students. In total, we analyzed ~30,000 surveys completed by ~4,800 students from 23 departments, covering 1,265 classes taught by 385 lecturers. We used both classic statistical and AI-based methods. Our findings suggest associations between high use of interactive learning methods and higher student evaluation scores, higher perceived effectiveness of distance learning, and clearer course teaching. A more interesting finding indicates that not only the extent of use, but also use of a variety of interactive learning methods significantly affects the perceived clarity of teaching and learning effectiveness. Based on the findings, we recommend that academic staff integrate a variety of interactive teaching methods, and especially short knowledge tests, in their courses (both online and frontal). Beyond these results, the prediction model we built can be used to examine what mix of different interactive learning methods might improve students’ evaluations of any given course.