Evaluating Prospects in Programming with Features in Graphic Function Use

Abstract

In this paper, we propose a quantitative evaluation method of students’ thinking in group learning. Thinking evaluation will become increasingly important in programming education in Japan. However, it is impossible for instructors to single-handedly evaluate their students’ thinking at the same time. It is necessary to provide a quantitative evaluation method that can be applied to a variety of educational situations in order to help instructors. We define coding vectors based on students’ source code that will serve as an indicator of evaluation. Moreover, we judge students’ prospects through a 3-step analysis with their coding vectors. We analyzed coding vectors for 22 participants obtained through a task experiment. We evaluated students’ thinking from three perspectives: visualization, distance, and direction. As a result, all three ways had the ability to grasp students’ thinking content. Coding vectors allow us to comprehensively judge students’ coding steps and their prospects. In this paper, we discuss the expressive power of coding vectors for coding content, and task settings appropriate for them.