Ac 2012-4551: Characterizing Students Handwritten Self-Explanations

Prior work has shown that self-explanation leads to greater learning gains and that those students who can clearly explain their solution process are more likely to generate good solutions. To our knowledge, no studies have measured the impact that self-explanation has on a student's solution process. In this work, we explore an unprecedented database containing digital records of the pen strokes from over 120 students' coursework over an entire quarter. This unique data allows for never before possible analyses of students' solution processes. In this paper, we compare the differences in class performance and solution processes between two groups of these students, one group that was required to supply handwritten self-explanations, another that was not. This comparison reveals that students who generate self-explanations along with their homework solutions perform better, learn core concepts better, and solve problems more like an expert than those students not required to generate self-explanation.