Revisiting UMBC's 6-Year Graduation Predictions

Resources

• Meeting recording (UMBC login req'd)
• Anonymous feedback

UMBC's Data Science team developed a predictive model to predict the likelihood a First Time Full Time Freshman would graduate in 6 years. The predictions are made after the student’s second and third semesters. Predictions are then compared to determine if a student’s graduation likelihood is increasing or decreasing from first year to second year.

In this session, DoIT's Robert Carpenter and Len Mancini will discuss how student predictions tend to change from semester 2 to semester 3, some of the observed trends from cohort to cohort and the implications of those trends.