Multiple choice questions to combat plagiarism and encourage conceptual learning

Abstract

Academic dishonesty (i.e., plagiarism or cheating in plain English) discredits and compromises the validity of University qualifications. All the effort an academic puts into the construction, administration and delivery of their courses is undermined if students are cheating. There are a number of studies describing methods for detecting plagiarism in written assignments. For example, there is software which can detect high levels of similarity between students’ work (Moeck 2002; Hamilton 2003), and there has been a special interest in identifying Internet-related cheating (Young 2001), commercial ghost-writing (Hammer 1976) and in alerting academics to other forms of electronically based plagiarism (McMurtry 2001; Heberling 2002). At first glance, these issues seem less relevant to our large Biochemistry and Molecular Biology classes (which have enrolments of between 200-700 students) because our assessment is based on laboratory work (including individualised practical skills tests) and end-of-semester examinations. However, not only is it clear that a great deal of cheating can occur in the examination room (especially when large numbers of multiple choice questions are employed), our experiences have also made us realise that the detection of cheating causes tremendous stress of itself. Thus the central theme of this paper is that it is a waste of time to employ sophisticated plagiarism detection solutions for in-semester written work if those assignments contribute relatively little to the final mark (as would be the case if the bulk of the final assessment is an easily corrupted multiple choice examination). Additionally, we believe that it is better to make multiple choice examinations that cannot be plagiarised than to burden oneself with the problem of dealing with suspected plagiarism. To this end we have developed examination solutions that prevent cheating and which, as a by-product, allows for flexibility in question design that facilitates the grading of individual options within multiple choice questions.