engleski

Post-test analysis of automatically generated multiple choice exams: a case study

Multiple choice exams (MCEs) are widely used to assess students' knowledge because they can be graded objectively, consistently, and fast. An aspect of MCEs often neglected is that they can also provide a valuable feedback to the teachers. MCE post-test analysis can be used to pinpoint invalid items or assess the clarity of taught topics. In this paper, we focus on the analysis of MCEs, elaborate on the kind of problems that can arise with MCEs, and how they can be detected. We describe our experiences gained on the Artificial intelligence course taught at our Faculty, and discuss the lessons learned.

knowledge assessment; multiple choice questions; post-test analysis; exam generation software

nije evidentirano