In this paper, we developed predictive models based on students’ reviewing behaviors in an Introductory Programming course. These patterns were captured using an educational technology that students used to review their graded paper- based assessments. Models were trained and tested with the goal of identifying students’ academic performance and those who might be in need of assistance. The results of the retrospective analysis show a reasonable accuracy. This suggests the possibility of developing interventions for students, such as providing feedback in the form of effective reviewing strategies.