Intelligent performance assessment in a virtual electronic laboratory

Abstract

Laboratory work, in the Undergraduate Engineering course, is aimed at enhancing students’ understanding of taught concepts and integrating theory and practice. This demands that laboratory work is synchronised with lectures in order to maximise its derivable learning outcomes, measurable through assessment. The typical high costs of traditional engineering laboratory, which often militates against its increased use and the synchronisation of laboratory and lectures, have, in addition to other factors, catalysed the increased adoption of virtual laboratories as a complement to the traditional engineering laboratory. In extreme cases, virtual laboratories could serve as alternative means of providing, albeit simulated, meaningful practical experiences. A Virtual Electronic Laboratory (VEL), which can be used to undertake a range of undergraduate electronic engineering curriculum-based laboratory activities, in a realistic manner, has been implemented as part of the work presented in this thesis. The VEL incorporates a Bayesian Network (BN)-based model for the performance assessment of students’ laboratory work in the VEL. Detailed descriptions of the VEL and the assessment model are given. The evaluation of the entire system is in two phases: evaluation of the VEL as a tool for facilitating students’ deeper understanding of fundamental engineering concepts taught in lectures; and evaluation of the assessment model within the context of the VEL environment. The VEL is evaluated at two different engineering faculties, in two separate universities. Results from the evaluation of the VEL show the effectiveness of the VEL to enhance students’ learning, in the light of appropriate learning scenarios, and provide evidence and support for the use of virtual laboratories in the engineering educational context. Performance data, extracted from students’ behaviour logs (captured and recorded during the evaluation of the VEL) are used to evaluate the assessment model. Results of the evaluation demonstrate the effectiveness of the model as an assessment tool, and the practicability of the performance assessment of students’ laboratory work from their observed behaviour in a virtual learning environment.

Date of Award	Apr 2011
Original language	English
Awarding Institution	University of Portsmouth
Supervisor	Djamel Azzi (Supervisor)