Building Data Science Capabilities into University Data Warehouse to Predict Graduation

TL;DR

This paper describes integrating data science capabilities into a university data warehouse to predict student graduation outcomes, involving infrastructure enhancements and a pilot study using student registry data.

Contribution

It introduces a novel approach to embed predictive analytics within university data infrastructure for real-time student outcome predictions.

Findings

Successful integration of data science lab into data warehouse

Pilot model predicting graduation probability and time-to-degree

Identified ethical and legal considerations for deployment

Abstract

The discipline of data science emerged to combine statistical methods with computing. At Aalto University, Finland, we have taken first steps to bring educational data science as a part of daily operations of Management Information Services. This required changes in IT environment: we enhanced data warehouse infrastructure with a data science lab, where we can read predictive model training data from data warehouse database and use the created predictive models in database queries. We then conducted a data science pilot with an objective to predict students' graduation probability and time-to-degree with student registry data. Further ethical and legal considerations are needed before using predictions in daily operations of the university.