It is, of course, a serious academic offence to pass off somebody else's work as your own. Universities are in the business of accrediting students as having acquired certain knowledge and skills, and it is a fraud on potential employers if students have cheated in their assessments. Yet research suggests that plagiarism is much more widespread than most academics realise, to the extent that degree standards are in serious danger.

University administrations have responded to the problem of plagiarism by treating it as a crime. They issue dire warnings of heavy penalties for the offence; they use advanced methods of detection, such as Turnitin; and they have complex, quasi-judicial procedures for dealing with cases that are detected. However, the way students are taught and assessed is a far more significant factor in the occurrence of plagiarism than the criminal intentions of a small minority of students. If we are to improve the situation, we need to design the possibility of cheating out of the system, rather than focusing all our efforts on detection and punishment. There is no single change that can miraculously make plagiarism go away, but there are a number of things that can be done (and other things to be avoided) that together make plagiarism virtually unthinkable for students.

Central to any strategy for eliminating plagiarism is the need to foster a culture in which students are motivated to learn, and not merely to acquire a degree certificate by any means. Universities are fortunate in that most students have made a deliberate choice of the subject they want to study, even if some of them are motivated as much by the social life and a passport to a good job. As long as students want to learn, they know perfectly well that cheating does nothing to increase their knowledge. If a student has paid good money for guitar or driving lessons, they won't cheat, because they genuinely want to play the guitar for pleasure, or to drive a car competently. Their attitude to studying at university should be the same. But it is the responsibility of university teachers, from the moment students arrive, to provide a stimulating and rewarding environment in which they are conscious of making steady progress. Unfortunately, far too many students, particularly in their first year, spend most of their time sitting passively in lecture theatres, where they are expected to memorise what they are told (often covering ground already studied at A level) and regurgitate it at the end of the course. It is hardly surprising if they come to see academic life as little more than an exercise in jumping over hurdles.

It is a paradox that students have much more individual attention in their final year, when they are working on projects or dissertations, than in their first year. If the purpose of university education is to make students progressively more autonomous in their discipline, it should be the other way round. Students should have a personal relationship with their teachers throughout their career, and especially at the beginning. It is a fact of human nature that people are more honest in dealing with individuals they know personally than with impersonal institutions. People who fiddle expenses at work or fail to declare income to the taxman may be scrupulous about paying the right amount at the local corner shop. It is the same in university education. If students have a personal relationship with a teacher they perceive as helping them to learn, they won't cheat; but if they are required to submit work to an impersonal institution, as anonymous ciphers, then morality may fly out the window.

É«ºÐÖ±²¥

Plagiarism is possible only when the learning and assessment structure allows extraneous material to be passed off as a student's own work. As long as students are assessed simply on recall of factual information, there is no way of telling where the information has come from. It is a long-standing joke that students are severely punished for plagiarism, except when they plagiarise their own teachers. But the serious message of this joke is that there is something amiss if students can make free with the intellectual property of their teachers but not with that of anyone else. Students pay fees to be taught, not to buy the product of their teachers' research. What students get from their teachers in lectures or course handouts should be treated in exactly the same way as any other secondary literature. And if a properly referenced assignment comes out as nothing more than an amalgam of quotes from the teacher and other secondary sources, then it has no more value as a measure of the student's learning than the same document without the references (for which the student would be severely punished for plagiarism).

If copying others' work is to be avoided, students must be taught in such a way that they are actively engaged in their own learning, and assessed on what they can do as much as on what they know. They must be set tasks that they can complete only if they apply their own thinking, and give evidence of having done so. How this is to be done is discipline-specific. In my own experience as a teacher of the history of philosophy, I required students to give reasons for interpreting a difficult text one way rather than another, and for and against its adequacy as a philosophical doctrine as so interpreted. This information could not be downloaded from the internet or taken from printed materials. Students were forced to think for themselves, and plagiarism was never a serious problem.

É«ºÐÖ±²¥

A corollary of the need to make students learn actively is that the lecture should lose its central position in university teaching. If students are required to give answers to questions that have already been answered in lectures or course handouts, it is hardly surprising if they merely repeat what they have been taught. Teachers should set carefully structured tasks for students to perform, whether collectively or individually, and assess how well they achieve the outcomes. The educational focus will then be on active work in small groups rather than on passive attendance at lectures, and on what students can discover by their own efforts rather than on what they can remember from lectures.

One of the reasons why students are sometimes tempted to cut corners when assessed is because the stakes are so high. In most courses, a large proportion of assessed work counts towards the final degree result. This causes considerable stress, and students are reluctant to take the risk of thinking for themselves instead of borrowing the work of successful academics. It is usually the case that degree classification is arrived at by averaging far more marks than are necessary for a fair evaluation of student performance. The emphasis should be shifted radically from summative assessment, which does little to support student learning and encourages an unhealthy obsession with grades, to formative assessment, where teachers give personalised and relaxed advice on how to do better in future. It also has to be made clear to students that they are not assessed on the absolute value of the product (which, if downloaded from the internet, might well be better than they themselves are capable of), but rather on the evidence they provide of having used effective skills in arriving at a solution to a given problem.

Some may say that the kind of teaching and assessment I have outlined is all very well in an ideal world, but is unrealistic given the unfavourable student-staff ratios in many university departments. My response is that my suggestions do not necessarily imply an increase in workload (or if they do, then students are already being seriously short-changed for their tuition fees). Rather, they involve replacing one kind of activity with another: small-group work instead of lectures, and formative instead of summative assessment. Given that active involvement in a small group is widely considered to be a far more valuable educational experience than sitting in a lecture, students will not need as many timetabled hours as they currently spend in lectures, and it is better if a teacher is present only part of the time.

Finally, although I have presented my proposals as a solution to the problem of plagiarism, this is no more than a fortunate spin-off. There are much deeper educational reasons for moving away from a university culture centred on didactic lectures followed by summative assessment, to one in which the primary function of teachers is to guide and befriend students in their own active learning - the vision of great educationalists such as Immanuel Kant and Wilhelm von Humboldt.

How to beat the cheats

Ten years ago, people claimed that they were stopping plagiarism by using detection software. Now I commonly meet people who put their faith in assessment redesign. In both cases, I remind hopeful colleagues that dealing with a complex issue such as plagiarism requires many interconnected actions by teachers, institutions and, of course, students themselves.

But I am heartened by the shift from detection software to assessment redesign because assessment can capture students' time, shape their thinking and influence their decisions. Well, it can, but it is not inevitable; the key is redesigning assessment so that students cannot avoid the work of learning by copying, faking or paying someone else.

One tactic for making plagiarism less likely is to set tasks that do not have an "oven-ready" answer. You might ask for assignments to be submitted in a novel format (such as a radio play, a patient information leaflet or a mock submission to a parliamentary inquiry) rather than an essay. You could specify the application of a specific theory, or the use of a particular resource, thereby making it more difficult to recycle an essay bank document. By giving students individualised data, contexts, characteristics or situations, you lessen the chances of them copying from each other. Ask them to rank, justify or otherwise argue for an evidence-based solution and you are likely to be greeted with groans and complaints about "not finding anything". That groan may signal the recognition that they need to do some work. Result.

É«ºÐÖ±²¥

A second set of design strategies puts the focus on the process as well as the final product. You could observe one or more stages of production or verify designated stages by, for example, viewing drafts or setting up peer review between students. Factoring in demonstrations of students' learning through debates, presentations or in-class writing also reduces the chances of procrastination - students caught plagiarising often cite leaving things until the last minute as a reason for their decision to fake or buy another's work.

É«ºÐÖ±²¥

A third way to think about redesign is to focus on checking who actually did the work. One way to authenticate students' familiarity with their own work is by asking for a reflective commentary under observed conditions. Students could be asked: "In 10 minutes, here in class and referring to the essay you just handed in, list the three sources that were the most influential and say why." This provides a sample of a student's writing ability, albeit under time pressure and without benefit of spell-checkers and word processing. Some difference would be expected, but a significant one could ring alarm bells. Finally, a brief discussion or viva can reassure a teacher as to whose work they have been assessing. When students know this kind of authentication is planned, they might decide to do the work - or at least learn what it contains.

None of these initiatives is especially complex or problematic, although some require planning, and you might need to check regulations on what is permitted. Yet my experience is that teachers often struggle to apply these suggestions - and the reasons are surprisingly consistent.

The most common objection is that redesigning will take time: time to do it and more time to mark the results; you have to read individual, engaged pieces of student work rather than mechanically check them. Marking criteria for such assessments need to move beyond model answers and are more difficult to create. But complaints about the time it takes to create a task that gets students working seem exaggerated. Once teachers have a range of strategies, once they know how and when to use them, and if they pull in ideas and reactions from colleagues, then redesign takes a matter of minutes.

Another common concern is the risk of choosing the wrong approach for the context or course. If an assessment task must stay the same year on year (I am unconvinced that this is always necessary), it is still possible to use strategies for authentication.

Assessment redesign does become a struggle when teachers try to do it on their own, however. Academics who teach on the same programme and who review each other's assessment methods can exchange ideas, explore possibilities and challenge one another's methods.

Assessment redesign alone won't stop plagiarism, but the good news is that it is probably the strongest link in the interconnected activities needed to address this perennial problem.

Jude Carroll is an educational development consultant at Oxford Brookes University and author of A Handbook for Deterring Plagiarism in Higher Education (second edition, 2007).

Useful resources for plagiarism-unfriendly redesign:

É«ºÐÖ±²¥

.