Evaluation of alternative instructional strategies in intelligent coaching systems: an empirical study

Computer-based adaptive teaching can enhance individual learning effectiveness. Several researches have defined the concepts of Intelligent Coaching Systems and adaptive teaching in the past. However, empirical findings are limited to anecdotal evidence guiding the design of these systems. A prototype system based on the idea of adaptive teaching is built in Knowledge Pro which monitors a learning task. The learning task is on the topic of forecasting and forecasting models. The system monitors the student's performance and generate responses in the context of the student needs and student dialogue history. In this research the focus is on the expert evaluation and student performance modules. The expert evaluation module holds the declarative and procedural knowledge of domain concepts and the learning task performance process. The student performance module develops an overlay model of the student based on the expert evaluation module. The adaptive teaching module intervenes to minimize the gap between the student performance model and the expert evaluation model. The objectives of this research were to test: (1) the effectiveness of adaptive teaching and (2) the effectiveness of two instructional strategies concerning content in improving learning in a laboratory environment. The empirical results show that (1) adaptive teaching increases learning and learner satisfaction (2) instructions that provoke thought improve learning and satisfaction more than instructions that are specific to the problem and its solutions. The findings have the potential to enhance the effectiveness of teachers by creating learning environments that focus on the needs of individual students.