Beyond Surviving to Thriving: An Intelligent Assistance Performance Group Gap Analytics (PGGA) in a Gateway Course

• In Event: AI for Educators

Abstract

This study proposed a holistic analytical framework - Performance Group Gap Analytics (PGGA) - to analyze students’ learning from multiple levels and generate targeted insights to enhance the overall learning outcomes. Focusing on students’ learning performance in a course offered at a top 20 private mid-sized university, the PGGA framework incorporates three layers: performance group analysis layer, course assessment layer, and item analysis layer demonstrating the analysis from weekly assignment analysis to item level exam performance analysis. With this framework, we can exactly identify the topics and stages in the course occurring significant disparities in student performance. The analytical results indicate implications regarding better incorporating advanced analytical approaches as intelligent assistants rather than “black boxes” in educational settings.

Authors

• Bo Pei, University of South Florida

• G. Alex Ambrose, University of Notre Dame

• Ying Cheng, University of Notre Dame