posted on 2009-09-30, 13:47authored byDon Mackenzie
When testing higher level learning it is important to ensure that
e-assessment is as reliable and as rigorous as possible and that guessfactors
are reduced to the minimum in order to be confident that the final
score is a true reflection of the candidate’s ability.
A number of assessment systems, such as those provided by VLEs, provide
only basic controls on question behaviour and scoring together with relatively
simple ‘click & pick’ item types. For more advanced e-assessment it is
necessary to employ a dedicated e-assessment system that allows fine
control of how each item is delivered and flexible scoring methodologies that
allow for the award of partial credit for partially correct answers and stepwise
accreditation of the route to the answer even though the final answer may not
be correct.
History
School
University Academic and Administrative Support
Department
Professional Development
Research Unit
CAA Conference
Citation
MACKENZIE, D., 2008. The importance of controlling item behaviour and scoring methodology for testing higher level learning. IN: Khandia, F. (ed.). 12th CAA International Computer Assisted Assessment Conference : Proceedings of the Conference on 8th and 9th July 2008 at Loughborough University. Loughborough : Loughborough University, pp. 203-208