Evaluation of the Q-matrix Method of Fault Tolerant Knowledge Assessment PROCEEDINGS
Donald Bitzer, Tiffany Barnes, North Carolina State University, United States
E-Learn: World Conference on E-Learning in Corporate, Government, Healthcare, and Higher Education, in Montreal, Canada ISBN 978-1-880094-46-4 Publisher: Association for the Advancement of Computing in Education (AACE), Chesapeake, VA
Distance education, delivering high quality, individualized instruction in inexpensive, expandable, and accessible ways is a major challenge for today's universities. This paper presents the most recent work in the field of fault tolerant teaching (FTT). Fault tolerant teaching systems will augment computer-based learning materials with the tools to assess and correct student knowledge. FTT methods automatically create a concept model of an instructional topic, based solely on student responses to tutorial questions, using a statistical method related to data mining. This concept model can be used to assess student knowledge as he progresses through a tutorial, while tolerating the usual errors that occur during testing - such as a student guessing answers or mistyping a response. This paper presents the ongoing results of an experiment designed to test the validity and accuracy of FTT-based assessment methods.
Bitzer, D. & Barnes, T. (2002). Evaluation of the Q-matrix Method of Fault Tolerant Knowledge Assessment. In M. Driscoll & T. Reeves (Eds.), Proceedings of E-Learn 2002--World Conference on E-Learning in Corporate, Government, Healthcare, and Higher Education (pp. 1167-1170). Montreal, Canada: Association for the Advancement of Computing in Education (AACE). Retrieved November 25, 2017 from https://www.learntechlib.org/p/10358/.
© 2002 AACE