Comparison of Testing Evaluations between the Placement Test and the Final Examination
item response theory
It is considered to be difficult to design mathematics lectures for undergraduate classes if a variety of students exist in the same class. To grasp the structure of the classes as early as possible, placement tests are often performed. Using the testing results, students are classified into different level classes. However, the students in the lowest classes may lose pride and cooperation attitude. Thus, our university decided not to provide such classes. In such a situation, much more efforts are required for teachers to enhance students’ skills. To check if the progress of mathematics in undergraduate classes, we have compared the final test results with the placement test results. The item response theory (IRT) is used to evaluate the abilities of students rather than the classical test theory.
We have found that 1) students made progress by well-arranged lecture design, 2) there seems to be a positive correlation between the placement test results and the final test results in analysis classes, 3) much more progresses were seen for students without higher level high school education than those with that, 4) in linear algebra classes, no clear relationships were observed between the placement test results and the final test results. Above all, students made progress even though they were not educated well in high schools.
Departmental Bulletin Paper