Abstract

Originally designed for automated assessment in mathematics, the STACK assessment system is evolving to address the needs of a broader range of disciplines, including statistics and data science. This talk showcases recent advancements in STACK’s core design that enhance its ability to manage and scale with large datasets generated within questions. These improvements enable STACK to handle the complexity and volume typical of assessment in data-intensive fields.

This talk includes details on the enhancements made and the resultant computational efficiency gains in data handling. We also present examples of questions which demonstrate STACK’s new capabilities. These examples serve to highlight STACK’s potential to be applied in innovative ways to assessment tasks involving large datasets.