Efficient continuous online assessment of large classes

Continuous diagnostic assessment in engineering & mathematics subjects

Nathan Scott and Kevin Judd
University of Western Australia

Discipline Area

Engineering mechanics, calculus, statistics, linear algebra, thermodynamics, thermofluids

Key assessment issue addressed

Efficient continuous online assessment of large classes

The initiative

Our software platform is called "Jellyfish" and was developed at UWA by Judd and Scott.

Students solve a long series of problems on-line during the semester as their tutorial work and assessment. The problems in engineering have answers that are a number with units, and each student gets slightly randomised problems so their answers are different from their friends' answers. If a student enters a common wrong answer (that is, has made a classic mistake anticipated by staff) the computer system is set up to give specific feedback.

The maths extension of this system is much more sophisticated because the answers to problems take the form of an expression. Dr Kevin Judd of UWA Mathematics has programmed Mathematica(TM) to identify errors in student work and to provide very precise and detailed feedback. This software has been refined for several years and in a recent trial was found to trap 90% of all student input correctly - possibly the best anywhere in the world.

Each of the problem sets has an integral "forum" or bulletin board so students can ask questions about the problems on-line. All student questions and staff responses about a given problem are kept with the problem and this has proven to be very efficient. A given student query is answered ONCE by staff and the answer is then seen by all students.

The system has been used since 1995. Student surveys show a high level of approval and exam results are much the same as before 1995. However teaching time is down to about 30% of its former value.

The reason for the initiative's effectiveness

The system allows us to pose problems in a form that is appropriate to the discipline area, and the diagnostic feedback means that many common student queries are handled automatically.

Further details

Engineering: Dr Nathan Scott
Dept. of Mechanical & Materials Engineering
The University of Western Australia
nscott@mech.uwa.edu.au
http://www.mech.uwa.edu.au/nws/ae/

Maths: Dr Kevin Judd
Dept. of Mathematics & Statistics
The University of Western Australia
kevin@maths.uwa.edu.au
http://calmaeth.maths.uwa.edu.au/


 
 

Back to Top

 

This work is copyright. It may be reproduced in whole or in part for study or training purposes subject to the inclusion of the source and no commercial usage or sale. Reproduction for purposes other than those indicated above, require the written permission from the Commonwealth available through AusInfo.• © Copyright 2002, AUTC.