It is inevitable that waves will rise and fall. What is not known is the form each will take, at least until they begin to take shape.
There is an emerging narrative that a switch from teacher evaluation to teacher development is required. The backlash against teacher evaluation began some years ago with calls to stop grading lesson observations and critiques of book scrutiny. These calls were justified on the basis of validity. The act of evaluating teachers’ performance was also dealt a blow by the fall from grace of progress data. Questions were raised about whether it was right to attribute progress so strongly to an individual teacher and if the data used to do so was an accurate measure of what children were learning.
But the intellectual demolishing of teacher evaluation does not lead easily to the dismantling of practices within schools. Teacher evaluation is alive and kicking across the country.
It is common for schools to take considerable time to drop old practices and adopt new ones. It may be that those that make policy do not encounter the emerging narrative, at least not sufficiently to be swept along by it. It may be that they find it difficult to accept that what they have invested so much time and energy in should now be abandoned. Or it could be that they don’t have a clear idea as to what they should do with their time instead. In many schools, teacher evaluation absorbs considerable management time and is the basis for many of the daily conversations between teachers and school leaders. It is the discourse that binds relationships together. If we stop talking about ‘standards’, what will we talk about?
Into this void comes teacher development. This emerging wave is given power by the usual influencers. Instructional coaching becomes the talk of the town on Twitter. The EEF publish a guidance report on effective professional development. The DfE talk about a ‘golden thread’ of professional development which will run through a teacher’s career, from a revised teacher training offer, to the Early Careers Framework, to restructured NPQs.
Of course this shift is welcomed by teachers. Who wouldn’t want to be ‘developed’ more than be ‘judged’?
But we know from studying previous shifts in the zeitgeist of schools that there will be unintended consequences and lethal mutations. These are hard to predict – if they weren’t, we would be avoiding them better. But that shouldn’t stop us trying, not because we want to put a downer on people’s faith in the next big thing, but because we want the effort that is about to be expended to have as many positive effects as possible.
My mind has been wandering around the question ‘what could possibly go wrong?‘ I keep coming back to the thought that the old furniture hasn’t been removed from the house before the new furniture is being moved in. What might be the consequence of this?
To answer this question, let’s remind ourselves of a fundamental misstep we often make when we seek to judge the quality of something, that is to move from description to prescription. I will call this the mimicking fallacy.
Reading through Dylan Wiliam’s 2014 publication for SSAT ‘Principled assessment design‘ last week reminded me of this fallacy. Wiliam quotes Michael Polyani from over half a century ago when he talks about the tendency we have to make statements about what characterises quality work. We may think of these statements as quality descriptors. Polyani states that these statements have the appearance of rules, or maxims. The problem occurs when we move from attempting to describe what we observe in ‘high quality’ performances to prescribing action to those less adept at the operation in question. In other words, when we use descriptors of expert performance as maxims for others to follow.
Why is this a problem? Well, in many less complex spheres it wouldn’t be. Copying simple actions which lead to predictable effects is a good way of becoming more competent. However, teaching is an ‘expert pursuit’, which essentially means that it is too complex an endeavour to learn by copying those significantly more proficient than yourself. Highly effective teachers don’t just learn a set of simple behaviours which are known to have predictable effects. Indeed, the path to expertise may, counterintuitively, involve adopting different behaviours to the expert to begin with. It is for this reason that early career teachers need a different kind of development – perhaps more instructional coaching in specific ‘techniques’ – than later stage teachers who will be building more sophisticated mental models of practice.
I heard an interesting example which illustrates the problem of copying even relatively simple operations recently, but I cannot recall the source, so forgive me. In a series of experiments, researchers looked at how quickly a cat would learn to open a door by observing another cat doing it. Opening the door required the cat to adopt a small number of techniques and actions. First, the cat would bend their paw to form a hook. Second, they would extend their claws to get purchase. Third they would swipe at the edge of the door. But for the cat observing this operation, it was not at all clear which actions were necessary in opening the door – was it the motion, or the poise, or the use of claws? Therefore, when the observer-cat attempted to open the door they might swipe at the door without hooking or extending their claws, for instance, and the door would not open. It was only by observing ‘non-examples’ (i.e. failed attempts to open the door by cats only using some of the required actions) that the cat could infer the combination of actions required to be successful. In other words, the cat learned more by observing inexpert practise than expert practise. I suspect there was also a fair degree of trial and error involved.
Wiliam also quotes Robert Pirsig who suggested that maxims are post-hoc descriptions of quality rather than constituent definitions of it. The mimicking fallacy occurs when we forget this.
So, what does this have to do with the dangers of the emerging teacher development wave? My concern is that the legacy teacher evaluation practices will corrupt teacher development attempts as the maxims of expert teaching are applied to guide the development of practice. We know that new practices in schools are often overlayed onto old practices, rather than replacing them. In this instance, there is a significant risk that schools will continue to carry out teacher evaluation and adapt existing processes to make them look developmental. This would be damaging and wrong as a culture in which teacher development flourishes looks and feels entirely different to one which is geared towards teacher evaluation. If we don’t finish the job of dismantling the legacy infrastructure, the promise of teacher development will fail to be delivered.
We expect each new wave to sweep away the old one. But it doesn’t – not entirely. We are also repeatedly disappointed that new ideas which promise to be so much better than the old ideas don’t have the expected impact. It is conceivable that these two things are linked. Perhaps we should pay more attention to how the old bleeds into the new.
My other concern is that we risk thinking of teacher development as a silver bullet to the fundamental problems of school improvement. Resist, resist, resist. It is naive to suggest that ‘teacher development – good; teacher evaluation – bad’. It just isn’t anywhere near that simple. We may point to evidence that teacher evaluation has a ‘null effect’ (and that case is being built) but I predict that, in maybe 10 years from now, we will be finding similar null effects from all the effort we have invested in teacher development. And this will be for the same reason as for teacher evaluation. It will not be that it doesn’t work but that we haven’t made it work!
But that assertion needs more than a blog post to justify. For now, let’s proceed with caution and remember that when good ideas encounter reality they are at most risk of corruption.