Developing expertise #2 – Seek feedback

In my previous post I suggested the first step for teachers to develop expertise was to find ways to change the environment so that the feedback we get is unbiased. In this post we will consider why much of the feedback we do get is unhelpful and how to get more of the helpful stuff.

Many of the decisions we take  – in life generally and as teachers – are based not on reason and logic but on vague, nebulous feelings of ‘rightness’. Why did you buy the car you drive, the toothpaste you use, the shoes you wear? Why did you give Selma a detention for being late but let Robbie off with a warning? Why do you seen to get on so well with one class but struggle so much with another? Why did you ask that question in that way to that student? The answers, my friend, are drifting on the hidden breezes of consciousness. We may think we know why we behave as we do, but in very many cases we simply post-rationalise instinctive decisions to convince ourselves that we’re thinking, rational beings.

Analytical reasoning is hard, but intuitive decision making is effortless: we just do what ‘feels right’. To think through every decision we have to make when teaching a lesson would be impossible, even if it were desirable – which it probably isn’t – and so we trust our gut. The point is, our gut is only reliable if it gets good feedback on how effective its choices are.

Some of the feedback we get is excellent. When we make the decision to give a detention to Selma and not to Robbie we find out whether or not our judgement was correct. If both students subsequently start arriving to lessons on time then clearly our choice was justified. But how did we know? If we’ve been teaching any length of time we’ll have instinctively recognised that Selma’s and Robbie’s behaviour fit a template we recognised. We will probably have tried a range of options with both types of student in the past and will have developed an intuitive sense of what is likely to work in the future. We don’t sit and work all this out – that would take way too long – we just recognise the patterns and make out decision. Gary Klein’s research into ‘naturalistic decision making‘ has shown that experienced fire chiefs rarely if ever weigh up options to select the best way to put out a fire, instead they recognise feature of a fire and the ways that have worked in the past to extinguish similar conflagrations.

While getting students to turn up to lesson on time is important part of effective teaching, it’s not the most important part of the job. My view (and I accept this is contested) is that the role of teachers is to make students cleverer; to get them to know more about the world and become increasingly skilful at expressing their growing understandings. This is the area in which teachers are least likely to become better at.

Hamre et al (2009) indicated that while experience results in teachers become better and better at managing classrooms, there is a startling lack of development into teachers’ ‘instructional support’ over time.

Instructional support includes ‘concept development’, ‘quality of feedback’ and ‘language modelling’. These three aspects of teaching, how well we explain new ideas, how well we support children in improving, and how well we get them to think in academic language, are arguably some of the most crucial aspects of what we do. Why then do we have a situation where it’s not routine for teachers to become better at them?

The problem is that if all we do is take feedback on whether our actions improve in-lesson performance, our intuitions about how best to help students retain and transfer what they’re learning elsewhere and later are unlikely to improve. In this situation we go through the same process of pattern recognition that we do when organising our classrooms, but our judgement will be faulty because the feedback we’ve received will have been biased. We’ll only know what makes children do better in the here and now.

For instance, if I’ve just taught you how to perform quadratic equations and I then give you a quadratic equation to solve what does the fact that you can solve it tell me about what you will be able to do next week? All too often, teachers take feedback from lessons that students have grasped a concept or mastered a skill and then move on. We miss the fact that we made it easy to perform well in the lesson by giving cues and prompts to support students and say things like, “Look back in your books,” or, “Talk to your partner,” or, “It’s on the board!” Instead we should predict that most students will forget most of what we taught and expect them to need regular re-teaching where cures and prompts are gradually taken away until the ability to perform quadratic equations, or whatever else we’re teaching, is firmly embedded in long-term memory.

The errors we make are compounded by the fact that we tend not to recognise our limitations and as we become more experienced, are likely become over-confident about our ability to teach effectively. This sort of over-confidence blinds us to other ways of working and deafens us to constructive input from others. Because we know our judgement is trustworthy in some areas, we believe it must be god in all areas.

What we need to develop ways of getting better at ‘intelligent sampling of outcomes’ over the longer term. How can we track retention and transfer next year and the year after that? If we can, we’ll get a much better idea of whether our teaching is genuinely effective.  In my last post I suggested starting lesson by finding out what students remember from previous lessons is a good idea:

My advice is to begin lessons with a short sequences of multiple choice questions (MCQs). One or two questions should relate to last lesson, one to last week, one to something learned the previous term and maybe one on a topic covered last year. The beauty of MCQs is that they can take up very little lesson time: simply display the question on the board, offer 4 plausible answers (preferably with at least one distractor being a common misconception) and get students to write A, B, C or D on a mini-whiteboard. You can see at a glance whether students have remembered the answers and students get to have their illusion of knowledge revealed; often they might think they know something only to have their ignorance revealed when asked a direct question. For the best results, there ought to be no stakes attached to these kinds of quizzes – the point is to remind students that what has been taught in previous lessons is still important and as an opportunity for you to collect feedback on whether your instruction is actually helping students learn more.

If you’re interested, here’s a couple of very useful posts from Joe Kirby on why to use MCQs and how to design better ones.

If we’re serious about developing expertise in teachers then it’s worth knowing that to improve the quality of teachers’ learning, we must first pay more explicit attention to feedback teachers could receive, and second, by considering what kinds of feedback they should seek in the first place.

The next post in this series will look at how we can impose ‘circuit breakers’ to avoid making mistakes.