12.10. Delaying and reducing feedback
What I used to think
I used to think – in fact, I used to know – two things about feedback:
Thus, when students handed homework in on a Friday, many a Sunday afternoon was spent marking students’ tests and homeworks, never daring to leave a cross or a question-mark on the page unless it was accompanied by a explanation, personalised comment, worked example and follow-up question, switching frantically between my red and green pens. I used to dread the weekend.
Then Monday’s lesson would come around, and it was time for DIRT time. Students would be given their homework back, complete with reams of beautiful feedback, and within about three minutes I would wish I had never bothered. Students either told me they had finished everything, or claimed they didn’t have a clue what to do.
Sources of inspiration
My takeaway
A full discussion of what makes effective marking and feedback is beyond the scope of this book. For me, the bible on this matter is Dylan Wiliam’s Embedded Formative Assessment. However, I would like to briefly discuss one fascinating finding that has had a significant and positive impact on my teaching and (I think more importantly) on my work-life balance.
Hattie and Timperley (2007) explain that feedback is one of the most powerful influences on learning and achievement, but this impact can be either positive or negative. Given that the insistence by schools on written feedback is one of the key contributing factors to the workload crisis that is driving teachers out of our profession, we need to ensure that any time we do spend giving feedback is as beneficial to our students as possible.
When I interviewed Dylan Wiliam for my podcast, he uttered the immortal line: ‘The only good feedback is that which makes students think’ – a definite contender for tattoo number three. Despite the hours I put into my marking, my feedback did not make my students think. Ironically (or perhaps, unfortunately) the more detailed the feedback in terms of the information it provides the student, the less they need to think. By fully explaining where students have gone wrong and then providing scaffolded support in the follow-up question, students can pretty much breeze through DIRT time on autopilot.
One possibility to remedy this – and, indeed, it is the final of Bjork’s desirable difficulties that we will consider – is delaying and reducing feedback.
Soderstrom and Bjork (2015) present several studies that suggest that delaying or even reducing feedback can have a long-term benefit to students’ learning. Why? Well, because regular, immediate feedback can cause learners to become overly dependent upon it, perceiving it as a crutch to their learning. To relate this to themes covered in this chapter, immediate feedback prevents students from thinking hard and having the opportunity to start to forget. Like everything else in this chapter, delaying or reducing such feedback is likely to have a detrimental effect on short-term performance, but a positive effect on long-term learning.
But if we don’t identify student errors quickly, and provide our students with the means to understand and correct them, surely there is a danger that mistakes and misconceptions may become embedded?
Perhaps the important thing here is not necessarily the time delay between the answer and the feedback, but what happens in between. If students answer a question incorrectly, do not know they are incorrect, and then subsequently answer another 100 questions in that manner, then that misconception is likely to be difficult to remove – after all, practice makes permanent. However, if students answer just one question on a particular topic (and with my movement away from topic-based homework as described in this chapter, this is ever more likely), and are then compelled to reflect on their work following reduced feedback as described below, we may get the best of both worlds. Any negative effects may be negated, with the added bonus that students need to think more. This is pure speculation, but it makes sense combining the findings of several papers.
What I do now
When initially marking a piece of homework (or Monday’s low-stakes quiz), I now give no feedback whatsoever – I simply put ticks and crosses. Wiliam (2011) takes this a stage further, suggesting that instead of, say, putting ten ticks and five crosses, we instead simply inform the students that there are five mistakes for them to find and correct. We thus make feedback into detective work. I then give this work back to the students and see if they can identify the source of their errors and correct them. When I next take the books in, this is when I will give more detailed, task-focused feedback, again following the advice of Wiliam (2011).
I am trying to distinguish between three key reasons students may have got a homework question wrong:
I would argue that the kind of detailed feedback I was writing will only help in the first case. Students who have made a careless mistake may be alerted to that fact by the presence of a cross and be able to correct it; and students who could not be arsed will be no more or less arsed no matter what I write. Moreover, students who did not understand the question at the time have an opportunity to try again, reflect, self-explain or seek help from a peer during DIRT time. Either way, students are compelled to think.
There is another advantage to reducing and delaying such detailed feedback. Have you ever got half-way through a set of books, only to find you have been writing pretty much the same piece of (lengthy) feedback on most of your students’ attempts to a particular question? Good God, I know I have, and it is infuriating. By simply ticking and crossing I can identify patterns and trends much more quickly, and hence deal with them more effectively. For example, if most of my students have got a particular question wrong, I will probably choose to address it using example-problem pairs, Intelligent Practice, and so on, as opposed to writing the same thing in 30 books.
Directly related to this is a suggestion that Dylan Wiliam makes in Hendrick and Macpherson’s What Does This Look Like in the Classroom? (2017) which has revolutionised how I go through exam papers, particularly mocks. Instead of spending hours marking a set of papers, or spending a full lesson going through a paper – which may greatly benefit a student who has got 20% but not one who has got 90% – I now simply give students their papers back unmarked, sit them in small groups, give them a blank paper, and challenge them to come up with the best composite paper they can. As Wiliam explains, this technique ‘gives students retrieval practice when they actually do the test, takes advantage of the hypercorrection effect when students find out their answers were wrong, and also provides an opportunity for peer-tutoring (which, the research shows, benefits both those who receive, and those who give, help)’. It also helps us poor teachers have something resembling a life outside the classroom.