To free a person from error is to give, and not to take away.

Arthur Schopenhauer

In response to my last post, Cristina Milos pointed out that I use the term ‘feedback’ without providing any further clarification as to what I mean. She challenged me to explain exactly how I envisioned the feedback process taking place and to be clear about what, specifically, it ought to contain.

Now of course feedback can take various different forms, but seeing as I’ve been exploring ways to reduce teachers’ marking load, it’s probably apposite to address what written feedback might look like.

But, first some ground rules. The first principle I think’s important is that written feedback should result in more work for students than it does for teachers. Time wasting gimmicks of any kind – coloured pens, stickers, mandated structures – should be avoided. If it were possible to change the world by spending hours responding to every piece of work, it wouldn’t be sustainable and it wouldn’t be fair. Of course students are important; of course schools should strive for the best possible outcomes for their students, but not at the cost of staff well-being. Any school leader who argues otherwise is not someone I would ever willingly work for.

Second, written feedback should have some kind of meaningful impact on learning. If learning happens when we think hard, it should seek to provoke thought. It should provide hint and clues but make students work for ‘the answer’. As I’ve said many times before, by learning I mean the ability to retain information over the long-term and to transfer skills and ideas to new contexts. Now, I’ve no idea how you could hope to go about measuring this kind of impact and it’s going to be very hard to see in the short-term, but it makes sense to try.

Third, it seems sensible to suggest that as written feedback has a sort of permanence, it’s effects should aim to be permanent. Somewhat counter-intuitively, we need to stop marking work too frequently. Too much input from teachers results in what I’ve come to think of as ‘satnav feedback‘ – the sort of continuous, immediate feedback provided by a satellite navigation system – the sort of feedback that prevents memorisation and creates dependence. Ideally the sorts of interventions we offer should be carefully spaced out over a teaching sequence and only come after students have made some initial attempt to struggle on their own. For instance, when marking A level work I’d tell students I’d only mark each essay twice – once formatively, once summatively. When they handed it in for me to look at I’d ask them if they were sure they wanted to cash in one of their marking opportunities.

Fourth, while there’s nothing wrong per se with either giving grades or written praise, both seem to interfere with the ability to act on instructional feedback. If you really want to grade work, don’t then waste time writing feedback; if you’ve written feedback, don’t undermine yourself by slapping on a grade. If you really want to tell a student what went well, don’t then waste time how to make it even better. Praise this work and give feedback for improvement on that. Don’t spoil it by doing both at once.

So, with the ground rules in place, let’s see if we can work out what efficient, effective written feedback might look like. The first step is to stop marking for accuracy and pointing out students’ mistakes. Every time we point out and correct a spelling mistake, a missing capital letter or punctuation mark we undermine students’ ability to do this for themselves. My golden rule is that teachers should refuse to mark books which students have not proofread. Make proofreading for accuracy and improvement your minimum requirement; if they can’t be bothered to meet the minimum requirement, why on earth should you invest time marking their work? Use a simple proofreading code as a checklist and expect that all work is visibly annotated before it’s handed in. This is tough to do on your own but is very powerful juju if adopted as a school policy.

It’s also well worth asking students to highlight where they feel they most need your feedback. Expecting students to indicate where they might have encountered difficulties or taken a risk forces them to metacognitively engage with their own work and makes your marking much more focused.

With this minimum expectation met, I would then embark on marking. This involves making sure I’m happy with their proofreading. If they’ve not identified an error then I would use my knowledge of the student to judge whether it’s worth me pointing it out now or just logging that they’ve made the mistake. If they’ve missed something I know they know, then I’d return the book unmarked and get them to correct their work in their own time. The pressure to avoid careless mistakes should be so great that it starts to seem easier to get it right first time.

As far as I can see, most of the value in marking comes from reading students’ work. The process of writing out comments is laborious, repetitive and time-consuming: avoid it. If you’re an experienced teacher you can pretty well anticipate the kinds of comment you’ll end up having to write in books even before you pick up your red pen. My advice would be to make a list of your predictions and assign each a number. When you encounter a situation where one of your pre-prepared comments needs applying, simply write the assigned number instead. Where possible it can help if the ‘comment’ is phrased as a question (this has the twin advantages of being more palatable for students to read and inviting them to consider possible answers: How could you…? Why did you…? Is there another way to…? etc.) If you come across a mistake or misconception you hadn’t anticipated, simply add it to the list. In this way, I was able to mark a set of 27 books in less than 10 minutes – all I was doing was reading and writing down numbers.

Then, when books were returned next lesson, the comments would be displayed along with the relevant number and students would be expected to write out ‘my’ comments themselves.

For example:

  1. Rewrite this sentence thinking about the effect of the verbs
  2. No more than 1 sentence starting with ‘I’ or ‘The’
  3. Rewrite sentences without ‘and’ or ‘but’
  4. Vary your sentence structure
  5. How else could you connect this paragraph?
  6. Suggest three other words you could use here

(Credit here must go to Joe Kirby from whom I adapted the idea of what he calls ‘symbol marking’.)

Cognitive load can be decreased by placing the number next to the mistake or increase it by putting them at the end of the piece or work as appropriate. The trick is to make your students work as hard as possible – I leave to your professional judgement to determine what’s possible for the students you teach.

As well as this saving me time, it also increased the likelihood students would process my comment as they were recording it. Finally, after writing out my comments they would have time to respond to them. As they got on with this, I would take the opportunity to talk to them about their work: “I was really pleased to see that you…” or “Can you tell me why you didn’t…?” or “I’m not happy about…” This space for conversation is the crucible where relationships are forged and epiphanies sometimes happen.

Is this the most effective way to mark books? Hell, how would I know? I’m a mainstream secondary English teacher and other a very patchy understanding of what might work best in other contexts. I really like the sound of Kev Lister’s RAG123 model (although I’ve never tried it) and I’m sure it offers at least as positive an outcome. But this way resulted in students working more than me, it made them metacognitively engage with and think about their work, and it made them produce work of an increasingly high standard. And it’s quick. If you have a different, perhaps better way to mark, then I’d be very interested to hear about it, but not if it takes significantly longer.

Whatever system you have for marking books, try applying these two questions to it:

  • How will I know if this marking will have an impact on students’ long-term retention and ability to transfer between contexts?
  • How could I have achieved this aim more efficiently?

If you’re interested in the evolution of my ideas, here are some other posts I’ve written about feedback and marking that have a bearing on all this:

Image courtesy of Shutterstock