posted on 2006-05-24, 14:51authored byMariana Lilley, Trevor Barker
This paper reports on research undertaken at the University of Hertfordshire into the development and initial expert evaluation of a computer-adaptive testing programme based on Item Response Theory (IRT). The paper explains how the Three-Parameter Logistic model was implemented in the prototype. The underlying theory and assumptions of the model used in its development are also explained, along with the limitations and benefits of the computer-adaptive test (CAT) approach compared to traditional computer-based test (CBT) methods. In this paper use of the prototype as an alternative to the current method used by the University is evaluated by experts, and summaries of their reports and recommendations are presented. This paper also describes plans for developing this work further, including its use in computer-based student modelling where an accurate estimation of performance within a subject domain can be used to inform and adapt the choice of presentation of learning materials. Considerations for extending the CAT model to encompass other types of questions rather than multiple-choice or multiple-response questions are also presented.
History
School
University Academic and Administrative Support
Department
Professional Development
Research Unit
CAA Conference
Pages
314692 bytes
Citation
LILLEY, M. and BARKER, T., 2002. The Development and Evaluation of a Computer-Adaptive Testing Application for English Language. IN: Proceedings of the 6th CAA Conference, Loughborough: Loughborough University