A Study on Parallel Test Sheet Composition Algorithm for Multiple Assessment Requirements

• Main Authors: Ji-Yu Lin, 林繼昱
• Other Authors: Gwo-Jen Hwang
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/16168799920104617689

碩士 === 國立暨南國際大學 === 資訊管理學系 === 95 === In recent years, information technology has been widely applied to various educational applications. As test is the most effective and popular way for evaluating the learning achievement of a student, computer-assisted testing has become a wide discussed issue. A well-constructed test sheet usually need to meet multiple assessment criteria, such as the average discrimination degree, difficulty degree, estimate testing time and the ratio of relevant concepts. Moreover, in a large scale and formal test, it is necessary to compose several test sheets with similar difficulty degree and discrimination degree, such that accident situations can be coped with; nevertheless, it is time-consuming to compose test sheet form very large item banks to meet multiple assessment criteria, not to mention the composition of multiple test sheets with similar difficulty degree and discrimination degree. To cope with this problem, a tabu search-based approach is proposed in this study. A series of experiments have been conducted to compare the efficiency and efficacy of this approach with other approaches. From the experimental results, it can be seen that the innovative approach is able to effectively and efficiently compose a set of identical test sheets from very large item banks while meeting multiple assessment criteria.