The Predictive Power of Ai-generated Formative Assessments on Students’ Academic Achievement in Chemistry

The cynosure of the study was to determine the potency of chemistry formative assessment generated by Natural Language Processing Artificial Intelligence tools in predicting students’ academic achievement in Chemistry. The study adopted the correlational research design. The direction of the study was provided by two research questions and two hypotheses. The study sampled 336 final-year chemistry-education undergraduate students in Enugu State, Nigeria. The sample was drawn by simple random sampling across public tertiary institutions in the state. The instruments for data collection were “AI-Generated Chemistry Test Questionnaire” (AIGCTQ) and a proforma of students’ achievement in Chemistry. The instruments were validated by three experts in the Department of Science Education, University of Nigeria, Nsukka. The reliability of AIGCTQ was 0.84. The data for the study was collected using google form and analyzed using Regression Analysis. Coefficient of determination was used for answering the research questions while the hypotheses were tested at 5% level of significance using regression t-test statistic. The findings of the study showed that AI-Generated Formative Assessment predicted 63.2% of students’ achievement in Chemistry. Also, gender significantly moderated the predictive power of AI-Generated Formative Assessment on students’ achievement in Chemistry. Based on the result, it was concluded that AI-generated formative assessments can be used to predict students’ achievement in chemistry. Hence, it was recommended among others that parents and teachers should encourage students to use AI-Generated formative assessment questions for predicting students’ academic achievement in Chemistry.