Educational Digital Twin
A new paradigm for modeling educational systems at scale, across time.
The Physical System
An educational system has components in the physical world — e.g., learners, teachers, institutions, grade levels, degrees, and policies — and it evolves through time.
To date, these physical components are often modeled in isolation and as static snapshots — for example, a university's course enrollment statistics captured at a single point in time.
The Virtual System
“A digital twin is a set of virtual information constructs that mimics the structure, context, and behavior of a natural, engineered, or social system (or system-of-systems), is dynamically updated with data from its physical twin, has a predictive capability, and informs decisions that realize value.” — NASEM 2023
We propose an approach toward an Educational Digital Twin — a set of constructs designed to dynamically reflect and evolve alongside an educational system.
Educational Digital Twin
At the core of this paradigm is a formal, semantic graph structure that organizes educational data. By applying graph-theoretic operations, this evolving model supports efficient querying, self-updating, and representation at multiple levels of granularity.
For example, at its finest granularity, the Educational Digital Twin graph models each student's academic readiness, enrollment and dropout decisions, course attempts, declared major, and earned degree at every time step. Yet it also supports rolling up to instantly simulate course pathways at the aggregate student level.
Future Applications
Educational digital twins make it possible to address pressing questions in education today — for example, detecting at-risk students early, simulating curriculum changes before they are deployed, and building adaptive learning systems that respond to a learner’s current state rather than their overall average performance.