Time to Revisit Existing Student’s Performance Evaluation Approach in Higher Education Sector in a New Era of ChatGPT — A Case Study

Abstract

AbstractArtificial intelligence-based tools are rapidly revolutionizing the field of higher education, yet to be explored in terms of their impact on existing higher education institutions’ (HEIs) practices adopted for continuous learning improvement, given the sparsity of the literature and empirical experiments in undergraduate degree programs. After the entry of ChatGPT -a conversational artificial intelligence (AI) tool that uses a deep learning model to generate human-like text response based on provided input—it has become crucial for HEIs to be exposed to the implications of AI-based tools on students’ learning outcomes, commonly measured using an assessment-based approach to improve program quality, teaching effectiveness, and other learning support. An empirical study has been conducted to test the ChatGPT capability of solving a variety of assignments (from different level courses of undergraduate degree programs) to compare its performance with the highest scored student(s). Further, the ChatGPT-generated assignments were tested using the best-known tools for plagiarism detection to determine whether they could pass the academic integrity tests, including Turnitin, GPTZero, and Copyleaks. The study reported the limitations of the Bot and highlighted the implications of the newly launched AI-based ChatGPT in academia, which calls for HEIs’ managers and regulators to revisit their existing practices used to monitor students’ learning progress and improve their educational programs.

Keywords

• ChatGpt
• performance evaluation
• learning outcomes
• academic integrity
• artificial intelligence