The Impact of AI on Educational Assessment: A Study of Adoption Rates, Performance Metrics and Learning Outcomes

Abstract

The Artificial Intelligence is the new valuation and assessment driver in the education segment whereby institutions are now pursuing this path to measure student learning and performances. This paper will analyze the current condition of AI implementation in the education assessment system, challenges, and opportunities that surround the implementation of AI. The study follows a mixed-method guideline according to which the analysis of more than 7000 students and teachers participating in the surveys in 16 countries was conducted. Moreover, a systematic literature review that consists of more than 150 published studies in the time frame of 2023–2025 was also included. We made a premise that AI-based assessment systems would optimize accuracy of assessment, give individualized feedback, learning and without jeopardizing academic integrity. The study design presupposed the quantitative investigation and study of the adoption rates, the performance measures and comparisons between the traditional and AI-supported evaluation processes. The findings reveal that 92 percent of the students are currently using AI tools in academic settings with 88 percent in assessment using generative AI in an unbelievable increase in 2024 (53 percent). It turns out that the opportunities do exist, e.g., the opportunity to increase graduation rates by 43 percent and the improvement of the individualized learning process, but there are also notable challenges, i.e., the issue of academic honesty, data privacy, and equity. It is claimed that the key to successful implementation of AI in the education evaluation process consists of the evolved policy framework, education of the teachers, and the balanced approach, which would utilise the AI potential and, at the same time, would save the rest of the human-centred evaluation values.

Ähnliche Arbeiten