Leveraging LLM/GAI Agents for Authentic Assessment in Competency-Based Online STEM Programs

Abstract

This research-to-practice full paper presents our innovative approach to integrate Large Language Models (LLMs) and Generative AI (GAI) into authentic assessment methodologies to enhance the accuracy, fairness, and timeliness of assessments in online Science, Technology, Engineering, and Mathematics (STEM) education programs using Competency-Based Learning (CBL). CBL, particularly suited for adult learners in fields like Computer Science and Information Technology, allows for personalized and practical learning experiences. However, challenges such as maintaining academic integrity and providing timely feedback persist. By leveraging LLM/ GAI-enabled agents, this study aims to address these challenges by offering automated, consistent feedback, creating realistic practice scenarios, and facilitating peer reviews. The research also examines the effectiveness of LLM/GAI-enabled agents in maintaining academic integrity and delivering personalized feedback. The findings can contribute to developing more effective and fair assessment methodologies, ultimately improving learning outcomes in online STEM education programs.

Ähnliche Arbeiten