Senior Research Associate George Mason University Fairfax, Virginia, United States
Session Abstract: Markov Chain models have been used to predict the movement of students as they progress from first year to second year and beyond. Yet, students move through their studies in many ways (e.g., changing programs, shifting to part-time, moving to in-state status). This session will demonstrate how Markov Chain models can be enhanced to account for these additional movements and predict more granular counts of returning students. For example, instead of projecting the number of students in the engineering department who will return in following years, the enhanced model can account for students who move in from other departments, producing more accurate unit-level forecasts that can be used to make effective budgetary and programmatic decisions. The presenter will walk participants through the basics of a Markov Chain model using accessible terminology and showcase how to build a flexible model that captures the many ways students move through their studies.
