It’s a problem that has bedevilled schools and universities for years - how to manage students sitting the same exam across an extended period of time.

Now, new research has confirmed what many teachers suspected - sitting exams at different times creates “significant potential for cheating”.

A study has found that in one exam that was offered over a period of days, about one-in-20 students cheated.

Researchers at the University of Illinois in the United States looked at a large enrolment engineering course where an exam was offered over a period of days, with students able to choose when to sit it.

They found that some students opted to take it later so they could learn what was on the exam from those who have already taken it - a practice that the researchers called “collaborative cheating”.

According to their research, around 5 per cent of students studied in ways that indicated that they had been told information about what was on the exam.

Those students who engaged in collaborative cheating gained on average a 12 percentage point advantage for any questions where all students were given the same question, even if the question was modified in simple ways, like changing the numbers in a maths question.

Professor Craig Zilles, one of the researchers, said: “Even with class policies that indicate that students shouldn’t communicate about exams during the exam period, we know students do communicate.”

The researchers said that the solution to stopping this cheating was for question-setters to pick randomly from a pool of equally difficult questions for each question on the exam.

They found that having three to four questions to choose from in each slot of an exam was sufficient to make the advantage from collaborative cheating negligible.

This is because it is harder to get complete information about what questions are on the exam because a would-be cheater needs to talk to more fellow students, and because potential cheaters need to remember a lot more material.