A case study for the application of data and process mining in intervention program assessment and improvement

The University of Illinois at Chicago offers an intervention program by admitting students to its Honors College. We call this program Honors Program (HP). HP offers additional, valuable resources that can positively affect the educational trajectory of an individual student. The current selection process for admission into HP or dismissal from HP is traditional. Administration views a students’ current Cumulative Grade Point Average (CGPA) and does not look at the students’ past. In this paper, we take advantage of the educational history of a student to make the admission or dismissal of each student from HP more effective. We use data and process mining techniques to study students CGPA traces and their HP participation history. We measure the graduation rates of students based on their CGPA traces and HP participation history. We show that it is possible to improve the graduation rate of students if HP admission/dismissal rules are designed based on CGPA traces and HP participation history rather than just the current CGPA of students. The model produced from our study creates a method for both students and administration to evaluate whether or not a student benefits from participating in HP. For students who are eligible and might benefit from HP, the model also determines the optimal entering semester to HP.