A DATA-DRIVEN FLIPPED LEARNING MODEL FOR ENHANCED STUDENT OUTCOMES

This study investigates the integration of flipped classroom and learning analytics in the algorithm course through the Data-Driven Flipped Classroom model.  This research arises from the backdrop of the traditional teaching model in algorithmic courses, where a one-size-fits-all method was applied. The challenges appear as instructors encountered difficulties managing a class with diverse levels of assimilation, limiting the attainment of learning outcomes. Recognizing the need for a more adaptive and personalized model, the study introduces the Data Driven Flipped classroom model. This model provides a dynamic and personalized learning experience. Experimental research with computer engineering first year students at ESPRIT School of Engineering demonstrates the proposed model's effectiveness. Using this model, new learning activities were designed where we strategically employed pre-class Quizzes, in-class interventions, and post-class discussion forums to guide students through exercises of varying difficulty levels.  An experimental study was conducted to evaluate students' results and impressions. Survey results of the students (N=60) who participated in the experiment (experimental group) were compared to the results of the students from the control group (N=60). Significant improvements in students' problem-solving abilities, especially among those with lower starting assessments, demonstrate the model's potential to alter education.