An adaptive assessment system to compose serial test sheets using item response theory

Abstract

The conventional learning environment is being rapidly supplemented by an E-Learning environment, particularly Computer Assisted Instruction (CAI). Each learner has different learning status and therefore should use different test items in their evaluation. The Computerized Adaptive Test (CAT) can adjust the degree of difficulty of test items dynamically depending on their ability. One of the most important and challenging issues in conducting a good test is the construction of test sheets that can meet various criteria. Therefore, several measures have been proposed here to represent the quality of each test item. In E-Learning, there is a critical need of formative assessment and hence series of test sheets to be composed which pose some challenges. Consequently the selection of appropriate test items is important when constructing an assessment sheet that meets multi-criteria assessment requirements, such as expected difficulty degree, expected discrimination degree, specified distribution of relevant concept weights and exposure frequency of items. All behavior like selection of items, evaluation and termination of test are based on Item Response Theory, an educational theory. An adaptive assessment system is proposed to compose serial test sheets. This approach will improve the efficiency of composing near optimal serial test items to meet multiple assessment criteria.