Visualizing Trends in Student Performance Across Computer Science Courses

Student retention is an important topic in Computer Science departments across the country. Keeping strong students and helping struggling students perform better are two fundamental components of improving retention. Isolating the cause(s) of students leaving the major is an important area of research. We endeavor to explore this problem using a visualization tool to probe student data within the beginning course sequence in Computer Science. We would like to see what patterns exist amongst students, focusing on success, failure, and repetition patterns throughout the first three courses. Identifying these patterns can help isolate some of the causes of decreased retention within the department, allowing us to address individual projects, courses, or exams that may be causing students exceptional difficulty or loss of interest. Due to the large amount of data and the variety of students' paths through their courses, it is essential that a visualization be developed to represent the data. Using graph layouts, parallel coordinates, color-mapping, and interactive selection, users can explore and query the data. Users can discover patterns within the data by selecting subgroups of students and examining the event sequences to find patterns of success, failure, and repetition amongst those students. Departments can use this information to isolate profiles of students for retention, remediation, and recruitment efforts as well as identify areas of the curriculum or instruction that can be improved.