AI-Driven Assessment in Iranian High School English Classes

This study explores the transformative impact of Artificial Intelligence (AI) in education, with a focus on the automatic generation of test questions. Generating test questions by human test developers is a time-consuming process, which can be facilitated by AI due to its efficacy in the field of assessment. This study analyzed the performance of 11th-grade Iranian high school students on a test of items generated through an Automated Item Generation (AIG) system, including both fill-in-the-blank and multiple-choice formats. It aimed to investigate the significance of the difference between student performance on these question formats. The results revealed a statistically significant difference favoring fill-in-the-blank questions, which underscores the students' superior performance in fill-in-the-blank items. These results provide insights for educators and test developers, emphasizing the potential for tailored assessment formats to enhance student performance.