Marking blind

When marking an essay, it can be better not to know who wrote it.

As a university teacher, one of my regular tasks is to mark assignments, and I want to be as fair as possible to the students. One method I use is to “mark blind,” namely without knowing the name of the student whose work I’m marking.

anonymous student

Most teachers try to be fair and say that knowing the identity of the student makes no difference to them. However, there’s plenty of research about in-group favouritism, where the in-group can be based on family, religion, age, ethnicity or viewpoints, among other possibilities. Teachers are likely to be affected by all sorts of unconscious bias, including expectations about how good a student is.

Students create impressions on their teachers. Some students are more articulate, engaging, humorous or astute in their comments. Then there are the effects of appearance, dress and demeanour. Maybe a student really tries hard, creating a favourable impression of diligence.

Wine tasters evaluate wines without knowing their origin. Vintages can vary considerably from year to year, so it’s better not to be influenced by previous perceptions. Similarly, the quality of a student’s work can vary from class to class and from assignment to assignment, and teacher expectations can affect evaluations.

Even a student’s name can influence perceptions. Male or female? Ethnic? Pretentious-sounding or ordinary? Stereotypes abound and can influence attitudes. If you know a student well, in-group favouritism is a risk; if not, then stereotype bias is a risk. When marking and knowing the student’s name, it’s therefore likely that some mental image of the student will be present, and it’s likely this has an effect on the marker.

A-plus

My way of limiting this bias is to mark assignments without knowing the student’s name. I ask students to list only their student number on the assignment, not their name.

f-minus

I find this changes my attitude while marking. My focus becomes to comment on the work done, with less concern about the relationship of the comments or mark to the student. I don’t worry about a student who comes across well in class doing poorly or about a seemingly lackadaisical student doing well. After finishing marking all the assignments, I go online to recombine student numbers with names, and send my comments to the students.

One good aspect of marking blind is that I can say honestly to students that my mark is on their work, not on them personally. They can be more confident that if they receive a good mark, it is a reflection of good work and likewise that if they receive a poor mark, it is not about who they are.

If only more essays

Practicalities

In recent years, I have had my students submit their work electronically, either directly to me by email or through an online forum. Usually this means I can see their names as well as their student numbers – but only temporarily. I put the submitted files into a folder, with each file having as its name the student’s number. By the time there are a few files in the folder, I have forgotten which one is which.

Sometimes, when marking an assignment, I recognise the student. Perhaps it’s because we had talked about it beforehand, or the student gives some revealing personal detail, for example being from Finland when there’s only one Finnish student in the class. More commonly, though, it’s because the student includes their name somewhere in the assignment.

Whatever the reason, I put that assignment at the bottom of the pile and turn to another one. Usually after marking ten or so assignments, I’m on automatic pilot in terms of applying the assessment criteria, and knowing the student’s name is less important.

Sometimes, when marking an assignment, I think I know which student did it, but if I’m not absolutely sure, it helps me switch focus from who did the work to the quality of the work. I want the mark to be appropriate whoever did the work.

If I want to give feedback specifically for a student, supplementary to my comments on the student’s assignment, I can add this after reconnecting student names to assignments.

grading-2a

The pitfalls of familiarity

There’s an inherent tension in any system in which teachers mark their own students’ work. Teachers in such circumstances have two conflicting roles. One is to provide guidance, support and feedback to assist learning. The other is to provide an assessment of the student’s performance.

The trouble is that the assessment role can inhibit the support role. If students are worried about what mark they are going to get, they may be cautious about exposing their ignorance, thereby reducing opportunities for useful feedback. They may also try to curry favour with their teacher.

The way around this is to separate teaching and assessment roles. This occurs with research students in the Australian and British systems. The supervisor supports the student to produce a satisfactory thesis. Then the thesis is assessed by independent examiners. At the University of Wollongong, there are strict rules to ensure independence. At the PhD level, for example, examiners cannot have worked at the university in the past five years, nor have collaborated with any supervisor or the student, among other restrictions.

In years gone by, supervisors were examiners for their own students’ honours theses, but this was open to abuse, with some supervisors becoming advocates for their favoured students while some unfortunate students, who had clashed with their supervisors, were treated harshly. The rules were changed to prevent supervisors being examiners, though in some parts of the university there was resistance, with supervisors insisting that only they had the expertise to judge their students’ work.

One year, I made an arrangement with a colleague at another university: he would mark the final assignments from my undergraduate class and I’d do the same for him. This enabled me to be a support person for my students, giving them feedback on drafts before marking by my colleague. I thought the system was worthwhile, but it seems that few academics are receptive to this sort of exchange. My colleague never supplied me with the essays from his class. My guess is that he did not feel comfortable relinquishing his control over marks for his students. If, instead of needing to mark the work of 90 students in a semester, the figure was closer to 40, I might try again to arrange an exchange with a colleague or with one of the other tutors in my classes.

grading-2b

Other biases

Blind marking can limit biases due to knowing who did the work, but it doesn’t eliminate other sorts of biases. One of the most common is ideological: if students say things you agree with, they are more likely to create a favourable impression than if they challenge your beliefs. If you’re teaching on topics where there are strong differences in opinion, for example addressing abortion or biotechnology, being fair can be difficult.

There’s another problem too. Students are very sensitive to the views of their teachers, and many students will say what they think their teachers want to hear. This is probably more damaging and insidious than teacher bias itself.

Many years ago, I taught a course on environmental politics and used case studies as a basis for understanding theory. Many of the students were doing an environmental science degree, and most of the students thought of themselves as environmentally conscious, and it was hard to get them to think critically about their own beliefs. When nuclear power was the case study, nearly all students were opposed to it, and few students had the confidence to present pro-nuclear arguments. Furthermore, the students knew I was an opponent of nuclear power.

Then I introduced fluoridation as a case study. Some students asked me during class, “What do you think, Brian?” I’d respond that I was studying the controversy as a social scientist and didn’t have a strong personal opinion. This answer frustrated them: they obviously wanted to know my view so they would know better what to write in their assignments.

Furthermore, there was no standard environmental view about fluoridation, and different class members had different views on fluoridation, leading to more stimulating discussions than on other topics. The students had to think for themselves rather than regurgitate a standard line or say what they thought I wanted to hear.

On just one occasion, I used one of my books as a text. I didn’t like this, because I felt students were inhibited. Personally, I would have liked to hear their criticisms of my ideas, but few students have the confidence to question their teacher’s well-formed views. Basing teaching on your own research means you have greater knowledge, but does it help students learn more effectively?

Zits.20091222-grades

Conclusion

Fairness is just one consideration when marking. Ultimately, the goal is helping students to learn and to become independent, critical, ethical, self-motivated learners. How to do this is a continual challenge for which there is no single answer. I recommend trying blind marking to see what it’s like and to see how students respond.

Brian Martin
bmartin@uow.edu.au

Thanks to Anne Melano and Caroline Colton for useful comments.

Subject outlines illustrating how students can be instructed to submit blinded assignments.
CST228, 2015: see pages 11 and 15
BCM390, 2015: see page 17

See also: Marking essays: making it easier and more fun