Modeling Community College Pathways

Motivation

A student pathway is the longitudinal trajectory of attributes, actions, and outcomes that unfolds over an academic career — background preparedness, course selections, grades earned, enrollment gaps, and whether a student drops out, earns a degree, or transfers to a four-year institution. These pathways are especially complex at community colleges, where students arrive with widely varying levels of academic preparation, balance work and family obligations, and pursue goals ranging from workforce certificates to four-year degrees.

Overview

This project pursues two lines of work. The first is modeling these dynamic pathways at scale, across time. We propose the paradigm of an Educational Digital Twin: a set of information constructs that mimic the behavior of an educational system. At the core of this construct is a knowledge graph that organizes data into a semantic structure.

The second line of work tackles the challenges of data inaccessibility, which limits broader research and development. We develop a generative AI method that learns the statistical structure of real longitudinal pathways and produces realistic synthetic student data — enabling broader research and tool development without exposing individual records.

Impact

This work was supported in part by Texas Higher Education Coordinating Board (THECB) and Department of Energy grant DE-SC0021239. The opinions and conclusions expressed in this document are those of the authors and do not necessarily represent the opinions or policy of the THECB.