Adaptive testing system based on diagnosis mechanism and normal distribution

Abstract

In the distance learning environment, there are more and more teaching methods which are used via internet technology, such as computer-based teaching and computer-based testing. It not only reduces the correct time of instructors, but also gives students various changes in computer-based testing. It makes assessment not only a simple answer, but gives students more information about the tests. The computer adaptive testing includes the advantage of computer-based testing and also gives students different difficulties of testing items by student's ability level. This feature can not be replaced by traditional pencil and paper tests. Due to the various difficulties of testing items, we can understand the ability of students in class. In this paper, we proposed an adaptive testing system and used standard deviation to analyse student's test results. Teachers can understand the status of students during the test and also estimate student's ability in this system. Moreover, students can realise the learning problems of themselves via the analysed results.