Comparison of examination grades using item response theory: a case study

In item response theory (IRT), mathematical models are applied to analyze data from tests and questionnaires used to measure abilities, proficiency, personality traits and attitudes. This thesis is concerned with comparison of subjects, students and schools based on average examination grades using IRT. The difficulty of such comparisons is caused by student’s free choice of examination subjects. That means, students only sit examinations in subjects they have chosen themselves. However, if the students have different examination packages, their grades are probably not comparable. The main problem with using examination grades, or grade point averages (GPAs) is the incorrect assumption that all course grades mean essentially the same thing. However, there is always substantial variation among topics, courses, teachers, instructors and grading standards, so GPAs are not automatically comparable.