Double trouble for staff as student figures soar

A world without double marking would leave university lecturers free to concentrate on the important things in life such as reworking their lectures and researching their subjects, argues Alan Story, and the system, he insists, would be as fair as it ever was.

What if every civil and criminal case before British courts had to be tried simultaneously by two judges (or juries) in separate courtrooms before a verdict could be rendered? What would happen to the NHS if no doctor could treat a minor illness without receiving a second opinion? Both systems would become even more overburdened than at present.

Yet this is how the double-marking of student exams and essays functions: built into the decision-making about the mark for every student exam (and most essays) is the requirement of a second opinion. The consequences for the universities are the same: the system is over-stretched.

The double-marking system, however, just keeps lumbering along, seldom assessed itself, and seldom justified. Its logic is that two "blind" markers are required to ensure the "reliability" of examining and to check for potential mistaken judgments of colleagues - even though a sole teacher is trusted to teach a course, even though many factors more critically affect the "reliability" of grades than the number of markers involved, and even though many forms of summative assessment in current use have a scant relationship to education and teaching.

In my opinion, double-marking should be replaced at British universities (except for postgraduate dissertations and perhaps final-year undergraduate projects) by a single-marker system that is backed up by external examiners, and by a properly-functioning university appeal system to adjudicate student complaints. An appeal or second opinion on marks should be a right, as with a legal appeal, that is triggered by a complainant over a specific alleged injustice or mistake, not a mandatory prerequisite to the tens of thousands of marking decisions taken every year in every British university.

The time pressures resulting from double-marking are most pronounced at semesterised universities, such as the University of Hull, where I teach, because the volume of marking has now doubled, and become a twice- yearly ritual. I know colleagues here who now spend more than six weeks in a nine-month period solely marking student exams and essays.

But all British universities face similar pressures on time and resources. The mostly superfluous, labour-intensive, and time-consuming process of double-marking the academic work of sky-rocketing numbers of students is no longer a cost- or time-effective pedagogical activity.

Why should double-marking be phased out? First, its purported benefits are overblown and the exclusive focus on marker reliability/ unreliability often misses the point. Whether one or two markers (plus an external one) assess a given piece of work, numerous other factors affect the "reliability" or "objectivity" of a university assessment exercise. Research has revealed that asking different, but equally valid, questions of the same candidates on a second exam often results in widely different marks for individual candidates and a different overall mark profile - as does asking the same questions at a different time or place.

As studies have shown, perhaps the best way to reduce marker "error" is significantly to increase the number of questions asked on a given exam. But, when taken to the extreme, as with multiple choice questions, the examining exercise is dramatically changed, with undesirable and invalid inferences reached. Second, double-marking may, in fact, have negative effects and, again, be neither "reliable" nor "objective", its supposed advantage over alternatives.

Averaging the marks of two examiners tends to bunch marks much nearer to the mean. For example, both may not notice the particular brilliance or fresh insight that a single marker finds and wants to reward with a particularly good mark - and without having it averaged towards the mean. Marker "biases" against, for instance, poor handwriting, are just as likely to show up in the marks of two markers as with one. And if a given piece of work is marked by two examiners, who tend either to toughness or charitableness, the mark awarded will be the mean of "tough" or "charitable" marking. Subjective + subjective does not equal objective.

Studies have also shown that one of the major causes of marker unreliability is marker fatigue. Nothing causes more fatigue to markers than doubling their work load by double-marking.

Third, if we agree that both double- and single-marked papers may be "unreliable" - why else do we have external examiners? - let's look at the overall consequence to a student if either one or two markers assigns the so-called "wrong" mark in any one course out of the 10 to 20 courses on which a final degree classification is determined. Assume one marker (or two) awards a mark of 58 in that course, when a team of markers (and studies have shown that at least five markers are required to approach so-called "reliability") would have awarded a mark of 64. What could just as easily happen is that a single marker might award a 64 and the team a 58. In another course, marker drift may go in one or the other direction, favouring or penalising the student.

When you average out all the different combinations and permutations over 10 to 20 courses that are used to calculate the final degree class, the overall negative or positive consequences (for students) of single- marking these 10 to 20 courses are likely to be negligible.

In any event, with double marking, a "wrong" mark is often averaged with a second mark, hardly a way to ensure reliability.

The conclusion: double-marking clothes the inherently subjective process of examining with the aura of objectivity. Proponents of the current system say students would protest about any change. I am not so sure. If, by being freed from the rigours of double-marking, lecturers then had much more time to do really important tasks, such as providing well-thought- out and promptly returned comments on students' essays, reworking old lectures to make them more stimulating, developing new courses (without the dread of double-marking each student), experimenting with different types of formative assessment (the three-hour, closed book exam is not the only way to assess students!), and spending more time on research (which should reflect back in better quality lectures), I think students would welcome the change.

In Canada, where I was born and educated, a single marker system operates just as successfully and fairly as the British system. And I know one thing: if the double marker system were ever proposed at Canadian universities, there would be open rebellion in faculty ranks.

Single-markers of Britain, unite!

The writer is a lecturer at the University of Hull Law School.