Evaluating Prospects in Programming with Features in Graphic Function Use

Evaluation of thinking will become more important in Japanese programming tasks in the future. We propose a method for creating and analyzing quantitative features as an aid to evaluation in group learning. Based on the process of thinking activities in programming tasks, task experiments were conducted. Feature vectors were created from the data obtained in the experiment and visualized using principal component analysis. As a result, we discovered the possibility of identifying on the spot the condition of students with poor prospects. This will provide an unprecedented quantitative evaluation method for thinking activities during programming tasks.