Student Assessment

As an academic, one of the most important tasks I do is to mark students’ work. In terms of job satisfaction the task varies from those rare occasions when a student produces some superlative work to the usual drudgery of working through seemingly endless essays or exam scripts, each one a poor copy of the one before, and a sad representation of what is to come, especially if they are all answering the same questions. Some of the work is very good, some (not too many) dreadful, but much is – to be honest – just mediocre, neither good nor bad, but sufficiently competent to get a mid-range mark. Marking periods are just that, periods, preceded by pre-marking tension (PMT), and consisting of mood swings and the need for quantities of alcohol (which is difficult as I have given up drinking).

The main exception is usually when marking projects, where a lot of genuinely independent work has gone into solving a problem through scientific endeavour.

In the past marking consisted of big piles of paper, which gradually shifted from one side of the desk to the other, with various scribbles added, and ring stains of endless tea. Now it consists of long days of staring at a computer screen, trying to determine whether what is in front of you is a masterpiece or an essay bought from the large number of cheating sites that are available, where a student can purchase, for a reasonable sum, an essay written specially for the purpose, the standard varying according to cost, usually written by impecunious PhD students who have blurred lines about what is right and wrong and an empty food cupboard. I have no idea how common such cheating is, but for me it is a good reason to go back to exams, rather than coursework. More difficult to cheat at, but worse to mark because of the handwriting.

I did catch someone cheating in an exam once. I was walkingup and down the aisles and saw a student with a piece of paper covering whole pages of notes. I wanted to pick her up by her ear, yell at her and throw her outside, but in the modern world I took the cheat sheet and exam materials away, and led her out of the room, telling her that there would be trouble for cheating. Unfortunately people who cheat in this way rarely get thrown out, just given 0% and another chance. It would be better to birch them in public to discourage the others.

When marking was carried out on paper the world was a much easier place. Put a few comments (information for the student if if it is coursework or a second examiner for guidance), add a mark and you are done. Compile the marks at the end and hand them to an admistrator who will magically prepare them for the exam board. Now, with the advent of computer marking, the world should be an easier place because computers make the world easier don’t they? They don’t. Some reasons for marking now being harder because of computers:

• It is more difficult to read and flip backwards and forwards through an essay online than as a few sheets of paper
• It is fiddly to put comments on the script
• It is impossible to strike a big red line through something that annoys you
• While the marks should – you would think – automatically be entered into a system that produces a report for the exam board, it doesn’t. Often one has to rewrite the marks into some other system, eg Excel, and send that in to the administrator (who often doesn’t exist in the same way any more, but that is another story).
• Staring at a screen for hours, reading thousands and thousands of words is just not good for you
• etc, add your own, there is a limit to the length of a blog.

Second marking is a little controversial. It is, in principle, a good idea. Two people looking at a piece of work, coming up with independent marks and then coming to some agreement has to be a good thing. Or not. The problem with marking is that there is huge variation between markers. It is intrinsically subjective, no matter how much experience the marker has. There is ample evidence that markers vary, particularly at the extremes. Where one marker sees genius and awards a first class mark, another marker sees rubbish, and awards a low mark. Agreeing a compromise mark, sorry an objective mark, is all very well but there are power and other issues involved. A more senior academic is likely to get their view even though a PhD student doing the marking has perhaps been more conscientious. We also audit mark. This consists of one person marking properly and a second person looking at a sample of the scripts and going, ‘Yeah, OK, that’s fine’. Marking is subjective, we just hope it all comes out in the wash. Students have so many assessments that the final average should be just about right, we hope.

How do we deal with this? The solution, according to many people (who may or may not be employable in other industries) is to spend time on aims and learning outcomes. The aims of a module are something to do with what the module is meant to be about. The learning outcomes are what we hope the student is going to learn, and what we plan to assess. We then provide marking guidelines. These work at two levels, a) generally what we mean by A, B, C and fail; b) information about the content that should be included in the essay.

These are easy to deal with. The second one first. In my view, any academic who tells students what they expect to see in an essay is not really an academic. It is the student’s job to work out what to include and how, though the modern academic world is full of students being told to include this and that, but not the other. If a student asks me what should be in an assessment I just say I don’t know, because I don’t. I want the student to inspire me, teach me, inform me, educate me, not bore me stupid with rote answers that could be provided by a chimpanzee with an intellectual deficit.

The first one second. With marks, A means excellent, B means very good but a little flawed or lacking in spark, C means competent but mediocre (see above; remember, mediocre is not necessarily bad, just, well, mediocre). That makes sense, but in the modern world we are told to write long accounts of exactly what we mean by excellent, etc in our marking guidelines so that students can read them and somehow be helped, and markers can read them and somehow be helped. That is just bollocks or, thinking about it,if we already know that A means excellent how can we add anything to this? ‘The introduction needs to be excellent, the conclusion nees to be excellent’, etc. Just saying excellent means excellent is a pointless tautology, no matter how many words you attach to excellent.

There is also an argument about the amount of feedback we provide, from the odd tick and a mark through to the marker writing an essay about the essay. The general rule seems to be provide more feedback, and then provide even more. There are two useful forms of feedback. The best one, and the most important, is the mark. If you get 75% you know you have done very well, that your work is excellent. If you get 35% you know you have failed, which is bad. The other useful form of feedback is having a detailed discussion one on one between the student and marker and discussing the issues that arise from the essay. This can only happen if the marker has not provided guidelines which indicate what should and should not be in the essay, because in effect that is the marker’s essay and they are only marking how well their own work has been presented. It only works when the student has come up with an interesting answer that is a heuristic for further discussion. Most of the actual feedback we give (‘good point’, ‘needs clarifying’, ‘yes’, ‘need reference’, etc) is of little value. These are the things students should be able to understand themselves.

Some psychological research back in the 1970s explored the notion of experts and novices. It was argued, quite successfully I believe, that there are fundamental differences between novices, who have to express everything explicitly (remember learning to drive?) and experts, for whom knowledge and skills are implicit. If we take this to be true then as expert markers how can we create explicit marking guidelines if the skills of marking are implicit (which they are)?

The best way to move from the explicit to the implicit stage of marking, ie to become an expert, is to do the marking, and initially to share it with someone who has been marking for years. That is the best training. Oh yes, and when I started at a certain university I was advised to start your comments to the student with a positive remark (yes, it is possible to do this, ‘well done, you handed something in’), then explain why the work is crap/good, and then finish with another positive remark (‘You spelt your name right’). The classic sandwich.

While I could go on about this subject a long time, I have just one more point, perhaps the most important one, which I may come back to another time. Over the several decades I have been in academia standards have changed – dramatically, and for the worse. At the extreme an essay that might have got 58% in the past is now getting 72%, from 2.2 to 1st class. Certainly the standards have moved at least a full class. This is in some ways inevitable for two reasons, mass higher education which means that to ensure most people pass we have to lower standards because many of the people at university are now from the lower intellectual orders, basically the cohorts are not as bright as they were. Second, we have a customer focus. Students are not students any more, they are customers and they expect good marks. I will deal with this in detail in another blog.

In conclusion, marking can be exciting, but it is – let’s be honest – usually a chore. It is one of the less enjoyable parts of our job. I know students who have spent hours poring over their work may be disappointed by this but sorry, that is the way it is. Project work is usually the main exception. This blog is not meant as a whinge. As someone who has done a proper job, being an academic – including the marking – is still not real work, more of a lifestyle choice.