Artificial Intelligence-Generated Formative Assessment as a Predictor of Undergraduate Students’ Academic Achievement in Chemistry: A Case Study of Ebonyi State University, Nigeria.

Abstract

The rapid integration of Artificial Intelligence (AI) into higher education has significantly transformed instructional delivery and assessment practices. This study examined AI-generated formative assessment as a predictor of academic achievement in undergraduate Chemistry at Ebonyi State University. A correlational predictive design was employed. The sample comprised 214 second-year Chemistry students selected through stratified random sampling. AI-driven adaptive formative assessments were administered weekly over a 12-week semester via the university’s Learning Management System. Data were analyzed using descriptive statistics, Pearson Product Moment Correlation, and linear regression analysis. Results revealed a strong positive correlation between AI formative assessment scores and end-of-semester examination performance (r = 0.72, p < 0.001). Regression analysis showed that AI-generated formative assessment significantly predicted 52% of the variance in academic achievement (R² = 0.52, F(1,212) = 229.41, p < 0.001). The findings suggest that AI-based formative assessment systems provide robust predictive insights and can function as early academic risk detection tools. The study recommends institutional integration of AI-driven assessment platforms to enhance data-informed pedagogy and academic monitoring in Nigerian universities.