Teaching and Learning Forum 98 [ Contents ]

Computer-based teaching but pen-and-pencil examination?

Dr Karol Miller
Department of Mechanical and Materials Engineering
The University of Western Australia
The challenge of coordinating computer-based teaching with more conventional assessment has arisen in my teaching of Engineering Statistics for third year students at the University of Western Australia. The need for using computers for engineering and science education is usually unquestioned. One of the advantages which attracts me most is that computers allow to solve real size problems in front of students in a lecture theatre. Modern universities usually provide resources for computer-based teaching and learning. Lecturers have easy access to portable computers and projectors. Students have easy access to personal computers and common software. However, at present it seems to be beyond the financial capacity of most universities to conduct computer-based examinations for large groups of students. This would require having at least a hundred of fully serviced personal computers with engineering and scientific software. Therefore, at present, lecturers are faced with a difficult task of assessing computer-based taught students using pen-and-pencil exams.


Computers have infiltrated almost every aspect of our lives. We have used computers to prepare our papers as well as to submit them for this conference. Most of engineering students have attained some level of computer literacy a long time before they enter university. They will use computers on everyday basis during their studies and in their professional lives. Therefore, the need for computer-based teaching in engineering (and sciences) is usually unquestioned.

Graduates with professional degrees must possess certain, specified competencies. Lecturers' duty is not only to provide the students with opportunities to achieve these competencies but also to make sure that these skills have been actually acquired - and here comes the need for appropriate assessment techniques. In engineering many lecturers believe that only rigorous examinations can assure that the graduates acquired necessary proficiencies.

This paper poses an unanswered question - how to combine computer-based teaching with paper-and-pencil examinations - with hope that during the dilemma session the indication of the answer will emerge.

Advantages of computer-based teaching

The advantages of the use of computers in lecture theatres are caused by a few factors. The most important one is that simple large problems can be solved in a straightforward way in front of students' eyes. Modern mathematical symbolic manipulation packages, such as Mathematica (Wolfram, 1997), allow derivation of rather complicated equations in front of students. This feature is important to me because step by step derivations are disliked by the majority of engineering students. With Mathematica I am able to show the beginning of a derivation, indicate steps without showing intermediate results, and finally come up with a final expression - the one of interest in engineering applications. Additional advantage of using computers in a lecture theatre is the easiness of visualisation of different concepts. Almost every equation can be followed by a corresponding graphical representation. This is especially important in engineering education where most of the students consider themselves as practical people, not theoreticians.

In 1997 I was faced with a challenge of developing a teaching strategy for Engineering Statistics for third year students. Introductory Statistics is a very inviting subject to be taught with the help of computers. The underlying concepts are simple - they can be readily explained in a lecture theatre. Meaningful examples are usually large problems - but easily tractable in a lecture theatre with a help of a computer. The best examples of usefulness of computers can be found in teaching descriptive statistics. Let's consider lecturing about empirical probability density function (histogram). This function is usually used to make some sense from large collection of data. But how to show a meaningful example of a large data set without a computer? Consider, for example, lifetimes in hours of two hundred incandescent lamps (Ross, 1987). Such problem size is far beyond the reach of traditional lecturing style. A computer with appropriate software allows going through each step of construction of empirical probability density function, followed by appropriate visualisation, in about eleven minutes, including explanation.

Assessment strategies

In engineering lecturers are usually faced by large classes. My Engineering Statistics is attended by 78 students. This has an impact on the choice of assessment techniques. Engineering Course at the University of Western Australia offers an accredited degree. One of the duties of lecturers is assuring that the graduates acquired prescribed competencies. This poses additional restrictions on the assessment techniques - in most of the units the major component of the mark can not be awarded basing on assignments and projects completed at home. Traditional examinations, having al least 80% weight towards the final mark, prevail. In accord with this tendency, Engineering Statistics is currently being assessed basing on the final examination only.

Not many universities could afford to maintain an examination venue equipped with a hundred, fully operational and serviced, up to date personal computers with typical engineering and scientific software installed. Currently it is not possible in the Department of Mechanical and Materials Engineering of the University of Western Australia. However, we provide about fifty modern PC Pentium computers for our undergraduate students. Therefore computer-based teaching and learning is perfectly possible.


Computer-based teaching and learning in Engineering has a lot of advantages over a more traditional, blackboard or transparencies based approach. However, for economic reasons, it is not possible to conduct computer-based examinations for large groups of students. Lecturers in Engineering and Sciences are faced with a dilemma: computer-based teaching and learning but pen-and-pencil examinations?


The author would like to thank Professors James Trevelyan and John Wager for many useful comments and help in developing teaching strategy in Engineering Statistics.


Ross, S. M. (1987). Introduction to Probability and Statistics. John Wiley & Sons, New York, 1987, p.147.

Wolfram, S. (1997). Mathematica Book, 3rd ed., Wolfram Media, Cambridge University Press, USA.

Please cite as: Miller, K. (1998). Computer-based teaching but pen-and-pencil examination? In Black, B. and Stanley, N. (Eds), Teaching and Learning in Changing Times, 223-224. Proceedings of the 7th Annual Teaching Learning Forum, The University of Western Australia, February 1998. Perth: UWA. http://lsn.curtin.edu.au/tlf/tlf1998/miller.html

[ TL Forum 1998 Proceedings Contents ] [ TL Forums Index ]
HTML: Roger Atkinson, Teaching and Learning Centre, Murdoch University [rjatkinson@bigpond.com]
This URL: http://lsn.curtin.edu.au/tlf/tlf1998/miller.html
Last revision: 10 Mar 2002. The University of Western Australia
Previous URL 31 Dec 1997 to 10 Mar 2002 http://cleo.murdoch.edu.au/asu/pubs/tlf/tlf98/miller.html