Welcome to the InfoSci Platform
A Test Sheet Optimization Approach to Supporting Web-based Learning Diagnosis Using Group Testing Methods

A Test Sheet Optimization Approach to Supporting Web-based Learning Diagnosis Using Group Testing Methods

Chu-Fu Wang, Chih-Lung Lin, Gwo-Jen Hwang, Sheng-Pin Kung, Shin-Feng Chen
Copyright: © 2017 |Volume: 7 |Issue: 4 |Pages: 23
ISSN: 2155-6873|EISSN: 2155-6881|EISBN13: 9781522514480|DOI: 10.4018/IJOPCD.2017100101
Cite Article Cite Article

MLA

Wang, Chu-Fu, et al. "A Test Sheet Optimization Approach to Supporting Web-based Learning Diagnosis Using Group Testing Methods." IJOPCD vol.7, no.4 2017: pp.1-23. http://doi.org/10.4018/IJOPCD.2017100101

APA

Wang, C., Lin, C., Hwang, G., Kung, S., & Chen, S. (2017). A Test Sheet Optimization Approach to Supporting Web-based Learning Diagnosis Using Group Testing Methods. International Journal of Online Pedagogy and Course Design (IJOPCD), 7(4), 1-23. http://doi.org/10.4018/IJOPCD.2017100101

Chicago

Wang, Chu-Fu, et al. "A Test Sheet Optimization Approach to Supporting Web-based Learning Diagnosis Using Group Testing Methods," International Journal of Online Pedagogy and Course Design (IJOPCD) 7, no.4: 1-23. http://doi.org/10.4018/IJOPCD.2017100101

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Assessment can help teachers to examine the effectiveness of teaching and to diagnose the unfamiliar basic concepts (or attributes) of students within the testing scope. A web-based adaptive testing and diagnostic system can achieve the above objective efficiently and correctly. From a diagnostic point of view, the major concerns are to diagnose whether or not an examinee has learned each basic concept well in the testing scope, while also limiting the number of test items used (the testing length) to as few as possible, which will be directly related to the patience of the examinee. In this paper, we consider a test item selecting optimization diagnostic problem to reveal the mastery profile of an examinee (that is, to diagnose each basic concept's learning status (well learned/unfamiliar) in the testing scope) with a short testing length and a limited test item exposure rate. This paper uses the techniques of Group Testing theory for the design of our test item selecting algorithm. Two test item selecting strategies, the bisecting method and the doubling method, are proposed. The effectiveness of the proposed methods was evaluated by experimental simulations. The results show that both of the proposed algorithms use fewer test items and a limited test item exposure rate compared to the conventional methods.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.