Good afternoon everyone, you're listening to Radio EduTalk.

This is David Noble broadcasting live from the Burrell Room at the Marriott Hotel in Glasgow and delighted to have a very special guest this evening.

I'm going to welcome Dylan William who's, according to his Twitter biography, teacher, researcher and writer but also very well known as emeritus professor at the Institute of Education in London.

And tonight we're going to listen to Dylan talking around a couple of chapters in a recent book, a book that was published by SAGE this year called Assessment and Learning, second edition and the editor of that book is John Gardner.

So we're going to be looking at the two topics this evening are assessment for learning in the classroom and from teachers to schools, scaling up professional development for formative assessment.

As ever with Radio EduTalk you can get involved if you're listening to the live stream, you can email edutalker, that's edutalkr at hotmail.co.uk or if you're on Twitter if you just post your comment or indeed a question for Dylan and just include the hashtag edutalk, that's edutalk.

In terms of this show's topics you might also be interested in one of the chapters which is available on the SAGE website.

So if you're at a computer if you just go to bit.ly that's bit.ly forward slash then it's a capital X, a small q, capital C, small z, the number three and a small letter i, then that will open up chapter two that Dylan and Paul Black wrote for the aforementioned book.

So welcome to the show Dylan, I understand you're actually presently in Scotland?

I'm actually in Glasgow which is probably a couple hundred yards away from where you are.

Absolutely and you're working with a local authority this week?

I'm here in Glasgow doing work for the Tapestry Partnership so we'll go on to the work I've been doing with tapestry in the second part of the talk.

This is about how we bring effective classroom formative assessments to scale and so I can talk about that when we talk about the work we're doing on teacher learning communities.

Okay that's fantastic and if you wouldn't mind Dylan, just when we have guests on the show usually it's helpful if you could just give us a kind of brief introduction to your obviously very long and distinguished career in education before we start looking at the first topic assessment for learning in the classroom and which I understand is it was kind of written, the chapter was kind of written as a story so it'd be really great if you could you could take us through that in a moment but firstly if you could give us a flavour of your career to date.

Well I think the funny thing is that I never really intended to be a teacher.

I started out by actually wanting to be a rock musician and I went to university and started working in a band and then I realised that the band that I was working with needed money for equipment particularly a public address system so I was working semi professionally with these musicians and then started teaching to get money together for the band and did that for about two years and realised that I really couldn't carry on doing both it was just too impractical and to my shock and surprise I discovered I was actually enjoying the teaching more than I was enjoying working in the band so I gave up the band and I've never looked back I mean basically it's a decision I've never regretted and I then taught in a private school for one year in London schools then for seven years and then joined Chelsea College which then became part of King's College as a research fellow and then took over the running of the maths PGCE, the postgraduate certificate in education course there and then sort of moved up through the university hierarchy.

I never really understood how that happened but I ended up as dean of the school of education and then assistant principal of the college and then because I wasn't sure actually I think I probably was sure I didn't want a university administrator for the rest of my career I went to America for three years working for educational testing service as senior research director came back to the institute of education for four years but now for the last two years I've been freelance and I basically now spend my time half of each year in the US half of each year in the UK and basically I try to support teachers in improving their practice and the biggest part of that obviously is the focus on assessment for learning or formative assessment as we call it because we think that's the thing that teachers can do most effectively to improve their practice.

Thanks very much Dylan.

We've got about 40 minutes for the show this evening and as I've mentioned if listeners to the live stream wish to get in touch then I've mentioned how they can they can do so and I've also got a few questions that I'd be really grateful if I get the chance to put your way but if we maybe just start just now if you could just talk around the first topic this evening and assessment assessment for learning in the classroom.

Yeah well um I think this probably can be looked at as a series of traditional soviet style five-year plans.

I think back in about 1992-1993 Paul Black wrote a very important article on formative assessment in science and he was beginning to try to pull together research on formative assessment everybody thought it was a good thing but there wasn't that much evidence that it really had a big impact on student achievement and so from about 1993 to 1998 Paul Black and I worked sometimes independently sometimes together on formative assessment and drawing together the evidence that this was a an important part of teachers practice and this culminated in being asked to do a piece of work for the British Educational Research Association's policy task group on assessment.

They asked us to do a review of the literature and we did that by basically trawling through all the studies we could find that were remotely related to this.

We published our results in 1998 in a rather dense academic publication a 70-page journal article which somebody once described as a both elbows on the table read and because we thought it was actually quite important stuff we decided then to popularize that if you like by writing a booklet which drew out the policy implications and some ideas for practitioners and that was the thing that was called inside the black box and we were quite surprised by the reception but it did mainly because it was very short but people actually started picking up this and started talking about it and what we were doing in that booklet was to basically take the idea in the research study which is establishing that formative assessment is a good thing.

Students learn more when teachers do this but then to think about how we might apply it in practice so it was really going very much beyond the evidence given and so we in 1998 we published this and had some successes.

We then decided of course that the fact that this promised much wasn't really reliable unless we actually decided unless we showed how this worked in practice so from 1998 to 2003 we actually worked with a number of teachers to try and embed this in their practice and that phase of the work culminated in a publication that showed that even when you use standardized tests and examinations when teachers engage in these practices their students make more progress.

So if you think in one way that was kind of the end of the process but of course it wasn't really the end of the process because then we needed to get on to how we could actually do this at scale which we'll come on to later.

But I think looking back now we may have made a mistake even calling this stuff assessment because when people hear the word assessment they think of tests and quizzes and all the formal apparatus of assessment whereas in fact what the research that we've been looking at was really focusing on was the idea of assessment that happened as part of learning rather than that which happened at the end of it.

The other thing I think looking back is that we probably started in the wrong place.

The work we started with was very much focused in the area of feedback.

So we started by looking at feedback and then we found that you couldn't actually make sense of feedback without thinking about what kinds of questions students were responding to.

So we saw many examples where teachers weren't able to give feedback because they hadn't asked questions that really made the students thinking.

So very quickly our work on feedback expanded into work on questioning as well as feedback.

And then we realised that you couldn't really make sense of any of this without taking into account the role of the learner.

So feedback only works if learners pay attention to it and therefore involving the learner in this both as the recipient of feedback if you like but also learners as supports for each other, peer tutoring, things like that, complicated the mix even further.

So over time our conception of formative assessment grew from being something that we thought was just basically a way of thinking about feedback to being something much broader.

And so now we think about feedback as just one component of what we call formative assessment which involves the learners, their peers and the teachers all working together to improve learning.

And I'm interested to hear more Dylan about how you talk about the kind of changing nature of the teaching contract between the student or the pupil as we would call them in Scotland and the teacher.

And you seem to be suggesting that there has been in the cases that you've worked with through your research and that there needs to be more widely a change in the nature of dialogue between teacher and student.

I think that's right.

I think that obviously there are different, because this is to do with teachers, different teachers will be implementing this in different ways.

But I think what we saw in the most successful examples of this kind of practice was teachers becoming increasingly aware that what they were putting into the process was far less important than what the learners were getting out of it.

And so some people have described it as a transferring of responsibility from teacher to student.

But I think of it better as probably as a sharing of the responsibility.

So ultimately, the teacher is still responsible for making sure that students learn.

But of course, the teacher can't do that learning for the learner.

So what we I think we became clearer about was that the ultimate goal must be for teachers to help learners become autonomous in their own learning, to become in the psychological literature, self-regulated learners or self-regulating learners.

And so we saw more and more examples of teachers really empowering learners to take more control over their own learning.

And that really persuaded us, I think, that this was both desirable and possible.

Another thing that has happened subsequently is looking in primary schools.

We've been very impressed by how teachers have managed to involve students as young as four or five in their own learning in this way.

So it really is an extraordinarily powerful lesson for us, if you like, in seeing what practitioners can do with these ideas once they begin to adapt them into their practice.

And you reach the conclusion that feedback is part of, I think, classroom discourse, as you call it, needs to be a guide to improvement for the learning of the young person.

And you seem to be encouraging teachers, I get through empirical study, that that's more effective than referencing a young person and where they're at with their learning versus where they are referencing it against where they need to be in the teacher's eyes.

Is that correct?

Well, in a way, this is actually going back to the original definition of feedback, because feedback was a term that came from engineering systems theory.

And so originally feedback only functioned within a system.

You know, the classic example of a feedback system is the room thermostat.

You have a setting, the temperature you'd like the room to be.

You have a device for measuring the temperature in the room right now.

Then you have a way of comparing the desired temperature with the actual temperature.

And then, of course, you need a mechanism for changing the temperature of the room if the two aren't the same.

And so in engineering terms, feedback was always a component of a feedback system.

And what happened in psychology in the 1970s was that people started talking about any information given back to a learner as being feedback.

Now, that's not a concept that an engineer would recognise.

So in a way, I think one of the things we've been doing is putting the feedback back into a system.

So actually thinking about feedback as functioning within a learning system.

And then it becomes very clear that, for example, the most important feedback is that which is looked at.

If the students don't look at the feedback, it's not going to be effective.

And one of the things we've highlighted is if there's a comment and a score together, then what tends to happen is the students do not look at the comment.

They may not even notice the comment because they're so focused on the score.

So in a way, this is not very radical stuff.

It is just going back to the idea that feedback needs to function within a system.

And therefore, you think about the design of the system rather than thinking about the individual components.

In terms of making sure that the feedback is effective, then what kind of approaches have you suggested for teachers?

I think I read somewhere that one of the strategies would be to make sure that the feedback that's given to a young person can make some kind of demand of the next step for that young person in their learning, or some kind of response initiated or encouraged from them.

Yes, I mean, obviously, growing together such a vast literature into a simple set of ideas is not going to be in any way straightforward or even possible.

But I think one of the things that we've been very clear about is that feedback needs to cause a cognitive reaction rather than an emotional reaction.

Where feedback causes an emotional reaction, it is rarely effective.

I mean, there are a small number of students who are competitive, for example, and if you tell them that they're doing slightly less well than somebody against whom they're competing, then that emotional trigger may actually spur that student to greater effort.

But the vast majority of studies find that when you trigger an emotional reaction, then learning is much less likely.

So the first thing is to keep the emotions out of the situation.

For example, by making it difficult for students to compare themselves.

And teachers realise this, they don't actually get students to compete very often in terms of their sprint times or their high jump heights or their long jump distances.

They are much more likely to get students to focus on personal bests.

So in the same way that teachers of academic subjects can explore how they can actually report back to their students in terms of whether this work they just handed in is a personal best.

The important point is we keep the ego out of the situation.

And ideally, and this is a sort of summary of all the research, first is that feedback should cause thinking.

And secondly, that feedback should be more work for the recipient than the donor.

I often ask teachers, I asked teachers today, I was working with teachers in Renfrewshire, I asked them, does anybody here believe that your students spend as long reacting to your feedback as it takes you to provide it?

And not a single teacher did.

So generally, I think we give our students far too much feedback, but don't make them accountable for doing anything with it.

So the idea I think is that feedback should be less common and acted upon.

You know, ultimately marking is a very strange phenomenon in our education system.

Marking is really one to one tuition.

And as it's done currently, it's one to one tuition when the students aren't even listening.

You know, I think marking and that kind of feedback should be a last resort, because it's such an expensive process.

Teachers spending time commenting on individual students work, and then the students not doing anything with it.

So it's just about thinking about, if a teacher's time is very constrained, and therefore, how do you make sure that the time that is being spent is being spent in a way that's as productive as possible.

That's really been a driver for all this work.

And later on, we're going to talk about teacher learning communities and I suppose, improving professional development so that the key messages are embedded in not just individual teachers in a school, but right across schools.

And I'm just wondering how that ties in with, you know, on the horizon in Scotland, Dylan, I'm sure you'll be aware, you know, we've got a new suite of professional standards.

We've got a system of professional update on the horizon, and we're trying to improve systems of professional performance review and development for teachers working with managers and leaders in schools.

And it was interesting, I noticed in the chapter you talk about accountability pressures as kind of limiting progress in terms of formative assessment.

And you also say that much feedback, I'm talking obviously teacher to student here, is social or managerial.

And I just wonder to what extent that plays out in terms of the relationship between a teacher and their manager.

And you say that, obviously, feedback needs to be entirely about learning.

So I wonder how that will play out in the future.

Well, I think that what we're discovering in just about every context in which it's been studied is that summative uses of assessment drive out formative functions.

So many schools try to make their accountability assessments of teachers, both summative and formative, they always say that we have to monitor teachers practice or whatever.

But they also say, and of course, the idea is that teachers should improve as a result of this.

Now, the evidence is that this very rarely happens.

When people know that the rating they're going to get is going to be consequential in some way, then basically, the summative function drives out the formative function.

You know, because teachers are assessed, you're never going to find a teacher saying, I'm finding this idea, this teaching in this particular way, very difficult.

So what we like to do is to invite my senior manager in to come and watch me do this thing that I may not be able to do very well, to give me some advice.

That's simply not going to happen.

They're going to actually play safe, they're going to do the lesson, they know they're going to go really well, because the summative drives out the formative.

So obviously, we need to think about how we can get a constructive alignment between different factors in the system.

Now, achievement for students matters.

What I find interesting, and somewhat ironic, is that although the research evidence that we've produced suggests that there's nothing that schools can do that's going to have a bigger impact on student achievement than introduce more classroom formative assessment, people still see the idea of formative assessment as somehow in tension with improving student achievement.

You know, they say to me, I'd love to teach in this way, but I have to raise my exams grades.

And I think that's sad.

And I think the important point is that leaders have a role to play here in acting as cutouts.

So what typically happens is that governments dump on leaders, you know, in terms of demanding higher student achievement.

And the question is whether their leaders dump on teachers.

And what I'm saying is that I think what needs to happen in schools is that leaders need to act as cutouts.

They shouldn't just transmit the pressure that the government burdens them with onto teachers.

What they should do is to say, I know that we need to raise achievement.

I know that results are important for our young people.

But what we're going to do is we're going to invest in your professional learning.

So what I would like to see is actually an accountability framework, which worries far less about teachers' level of performance and worries far more about improvement.

We've learned that in student assessment for many years, that we used to believe that some students could learn and others could not.

And therefore, schooling was a process of sorting students into the educable and the ineducable.

And what we've realized is that if we actually organize learning for the benefit of all students, that all students can learn and actually achieve at high levels.

And what I'm convinced of in terms of my reading of the expertise research is that all teachers can achieve at high levels too.

And so the question is, what kind of system would support that?

My personal desire would be for any teacher who isn't actually so bad that they require intervention.

And there are teachers whose practice is pretty poor and they need some survival training, if you like.

They need help from a skilled mentor about what it is they need to fix about their teaching, just so that they are actually getting to a basic level of proficiency that is necessary just to get classrooms running smoothly.

But for any teacher who's not in that category, I think it should be up to each individual teacher to decide what to work on.

So, for example, my ideal system would involve each teacher meeting with a manager, maybe every six months, maybe every year.

And the manager would ask the teacher, what are your professional learning targets for this coming year?

And it would be up to the teacher to identify what it is that they wanted to work on over the coming year.

The assumption here is that the teacher is in a better position to judge what's going to improve their practice better than anybody else.

So then the teacher and the manager agree these targets.

And then the following year, six months later, they meet together to discuss progress towards this.

And again, it's the teacher's job to bring the evidence about the improvement to the meeting.

So it could be examples of video that show that the teachers have acquired new skills.

It could be the students' results on assessments or other kinds of performance tasks.

It could be interviews with students or questionnaire data from students showing that students have seen improvements in their own achievement as a result of changes the teacher has made.

But the important point is that it would be up to each individual teacher to improve their practice.

That would be the non-negotiable.

But it would be up to the teacher to decide what to work on.

And it would be up to the teacher to produce the evidence that the improvements that they planned to make did, in fact, materialise.

And Dylan, I know that in Chapter 2 of John Gardner's book Assessment and Learning, you mentioned that teachers require, just going back to formative assessment specifically, they need time, freedom and support to reflect and develop.

I'm sure that would be in terms of their practice around formative assessment as well as in other areas.

But you also say that during your research there was evidence of teachers adding and dropping components with regards to their practice around formative assessment.

A controversial question, but can teachers be trusted to make those decisions?

Can you say a bit more about why they were adding or dropping components to effective practice?

Was that just through taste or through their history in a teaching post, or was it in relation to feedback they were getting from young people as to which formative assessment approaches were working and which weren't?

Well, I think the basic idea was that each teacher has a certain number of things they need to do because those are school policies.

And so we envisaged a situation in which there was something that teachers could change about their practice without recourse to senior management of the school.

And the idea was that they expect to get on with that.

I mean, I do start from a somewhat generous assumption.

I think it's justified, actually.

I start from the assumption that all teachers are working as hard as they can right now.

And therefore, it is unreasonable to expect teachers to add things to their plate without first taking something else off.

And I think there's evidence for the reasonableness of that assumption in that when people add things to teachers' plates, when they add to the teacher's workload without taking anything else away, the things that are being added rarely get the attention they need to become effective additions to practice.

So I think there's both a moral and an empirical argument that the only way teachers can get better is to take things off their plates before they put things on that are worth doing.

So we did see teachers dropping stuff.

And we think, by and large, that was probably quite sensible.

Now, there are other times when what the teachers want to do actually cuts across the policies of the school, for example.

In many cases, teachers wanted to change the way they marked students' work, for example.

And what we've done there is to suggest to leaders that they might want to allow that kind of innovation to run for a limited period of time, just as a little experiment, and then for the teachers involved in that experiment to report back to the senior management about the experiment and how it went, so that they can see whether this is a worthwhile change to the policy for the whole school.

One of the interesting issues that comes up at this point, of course, is that people often tell me that the kinds of ideas I'm advocating only work well if every teacher in the school is doing it.

This is done as a whole school policy.

And I'm very happy to concede that these things almost always work better when everybody's doing it than when only a few other teachers involved are doing it.

But the worry I have with that kind of response is that it almost becomes an excuse not to do it, because I can't do it because nobody else in my school is doing it.

We've actually seen individual teachers making very effective changes in their own practice, even though nobody else in the school is doing that.

So for me, every teacher has a degree of professional autonomy, which I think they should use to improve their practice.

What I'm not going to do is to tell each teacher what they should be working on, because everything I know about teaching suggests that there is no formula, there is no kind of template that we can apply to teachers' practice.

I'm just constantly impressed by the ingenuity of teachers to solve problems.

And therefore, that's why I don't want to constrain that.

I want to put teachers in charge of their own professional learning, because I reckon that they are probably in a better position to judge what's going to work for their students than anybody else's.

And Dylan, your work in formative assessment and improving learning was back many years now.

And certainly, taking on board your excellent work has been a feature of professional development events, and I think just the professionalism of many teachers, since your first works were published with Paul Black.

So I'm just wondering, at this moment in time, it's difficult to generalise or to cast retros the whole of Scotland, but how do you feel it's going in terms of embedding formative assessment in Scotland's schools?

I mean, we had the support of the Scottish Executive many years ago, and it's still now mentioned within the Scottish Government.

So how would you appraise the current situation, and what kind of good practice have you seen down the years?

Well, I think that Scotland did a very good job of kicking off this process right at the end of the 1990s, the early 2000s.

So I think that the original focus was very welcome, the idea of assessment is for learning.

I think people got slightly seduced by the tips and techniques rather than thinking about this as being a vehicle for teacher learning.

So I think it got rather, and this may be inevitable for any innovation, it got rather packaged as being a thing that schools could do.

And many schools think they've done assessment is for learning, and so they're moving on to the next thing.

And so one of the challenges we've had is to realise that this idea of formative assessment is not just this year's innovation, it is actually the focus for every teacher's professional development for the rest of their careers.

Because ultimately, formative assessment focuses on the relationship between what did I do as a teacher, and what did my students learn?

What we do know from the empirical research is that what children learn as a result of what we do as teachers is basically unpredictable.

If it was predictable, we would never need to assess students, but it's not predictable, and therefore before we teach them anything else, we need to find out what the last bit of teaching we did caused them to learn.

And that's why that process of relating what did I do to what the students learn is such a powerful one.

It's the heart of reflective practice.

Assessment functions as the bridge between teaching and learning.

It's only through assessing that I can find out whether what I did as a teacher resulted in the learning that I wanted to get my students to engage in.

So I think we're having a second attempt working through the Tapestry Partnership to try and kick this off again as being not just another fad that we go through, but as a focus for teachers' professional development for as long as they remain in their posts.

And so in a way, we've spent a lot of time, we needed to spend a lot of time, telling people that we are now, you know, it's not a relaunch.

It's about saying, it's about going deeper into this process.

And we've worked with about half of the local authorities in Scotland now.

And so this work is being spread very effectively, and teachers are telling us it's the best professional development they've ever had, because it's putting them back in charge of their own professional learning.

So, you know, I think that a lot of the early stuff was a bit half-cocked.

We went off and focused on the tips and techniques.

And now we're actually having a more mature look at this and seeing what needs to be in place for teachers really to be taking charge of their own professional learning.

What kinds of information and knowledge do they need to make smart decisions about what to invest effort in?

And what kind of time do they need?

What kind of support do they need for that from their leaders for them to continue to develop their practice for as long as they remain in post?

And so I think that the work we're doing now is much more mature because we have a much healthier understanding, I think, of the difficulties of implementing these kinds of ideas in schools and how much support teachers need to do it.

You're listening to Dylan Wilton, who is, as we all know, an internationally renowned expert on learning and assessment, particularly formative assessment, or as you might know in Scotland, assessment is for learning.

We've got about 10 minutes left in today's show.

If you want to get in touch with a comment or question for Dylan, then you can email edutalker at hawthornmail.co.uk or if you're on Twitter you can use the hashtag edutalk.

The second chapter in John Gardner's book, or one of the other chapters in the book, is around from teachers to schools, scaling up professional development for formative assessment.

I know you've touched on the subject already, but would you like to say a bit more about how we can go about doing that and the role of teacher learning communities?

Well, I think the second chapter you're referring to, if you like, the next couple of episodes in the series of five-year plans I mentioned.

From 1998 to 2003, we were working with teachers to implement this.

Paul Black and I were working face-to-face with teachers and we found that when we worked with them and supported them, their students got better achievement.

Obviously, neither Paul Black and I, nor anybody else, could easily get and work with every single teacher face-to-face.

So from 2003 to 2008, we worked on this idea of trying to scale this up and that's where we hit on this idea of teacher learning communities, the idea that we could set up self-help groups in schools to support each other in their learning.

One of the things we tried to do was to come up with a set of resources that any school could use to support their own professional learning.

I'm not naive enough to believe that a pack of materials could be posted to a school and lead to wonderful professional learning.

We had a great deal of effort by the school, but the point was that we wanted to make sure that we designed the package as if it was designed to stand on its own because if we didn't do that, then we'd probably make too many compromises in developing the ideas.

So from 2003 to 2008, we worked on a model that we could actually use to work across all schools and in particular, giving teachers a framework that was tight enough to point them in the right direction while at the same time being loose enough to allow individual teachers to find themselves and to exercise creativity and flexibility to adapt what's needed to their own classrooms.

So we developed this framework where we have five key elements of effective teacher learning, which is choice.

Every teacher is free to choose which of these kinds of ideas they want to explore for themselves.

Flexibility.

Teachers are encouraged to adapt these ideas to make them work in their own classrooms.

Teachers need to change these ideas.

While one teacher finds it works for them, it may not work without substantial modification from another teacher.

So the important idea is that teachers are expected to choose for themselves but then also to modify and adapt to make them effective for them.

The third element of the model is allowing teachers to take small steps.

We are now completely convinced that teacher expertise is very complex and comes from a substantial amount of repetition.

Therefore, the only way that teachers can become better is to develop new expertise.

That means being able to change slowly so that you actually get lots of practice at what you're changing.

So allowing teachers to progress as slowly as they like and not until they've got a new technique under their belt and it's really second nature to them to encourage them to move on to something else.

If you have to remember to do something, then it probably yet isn't second nature enough for it to be something that you're willing to do, that you can actually stop focusing on.

So the idea of small steps.

And then providing accountability and support.

Those are the last two aspects of the process.

You can think of these as two sides of the same coin.

Sometimes referred to as supportive accountability.

But the idea is that every teacher needs to make a commitment to improve their practice.

Not because they're not good enough, but because they can be even better.

And then the leaders support them in that process.

So these five elements of choice, flexibility, small steps, accountability and support have driven us towards this model of school-based teacher learning communities.

So that was, if you like, the end of the third five-year plan.

We were tempted to think that we'd done what we needed to do.

However, what we found since then is that even though we are very clear now that formative assessment is one of the most effective things you can do to help improve student outcomes, even though we're very clear that when teachers do this, students learn more, even though we're very clear that when teachers meet in these groups of 10 to 12 practitioners monthly to hold each other accountable for making changes, then teachers do make changes in their practice and their children learn more.

It turns out that the critical path actually has to go through the leaders of the schools.

And what we've currently discovered is that the hardest thing of all is getting the teachers the time to meet.

So time and time again, we've worked with schools where people have promised teachers time to meet and the time never materializes.

And so the thing is that this process of improving teacher capacity, which I think is the most important thing that any leader can do, just gets added to a long list of things that are also going on in the school.

So I think our current work is focusing on how can we get leaders to understand that actually improving the quality of teachers in their school is the number one job for leaders.

And if you're really serious about that as your number one job, then creating space for teacher learning should be actually hardwired, hardwritten into the timetable before anything else goes in.

So that's our current focus is really just how do we get over this idea that the most important job of the leaders of schools is to support the professional learning of the teachers in that school because that's where the value gets added and that's where the improvement takes place.

And it's very hard to get people to focus on a very small number of things.

Most schools have got far too many priorities and initiatives going on and the result is that when everything's a priority, nothing is.

And do you have any suggestions about how to actually go about constructing the teacher learning communities in terms of the background or positions of teachers?

You know, for example, I know that in the past, as I understand it, you thought of the vehicle for formative assessment being groups of subject specialists, secondary, and then as time has gone on, we've realised that our primary teachers have worked with and seen the benefits of working with younger children.

So, you know, should it be a secondary school clusters or groups of subject specialists within a, say, a large secondary school?

What should the make up be in primary?

Should it be literacy as one group and numeracy, health and wellbeing all feeding into, I suppose, a whole school formative assessment action plan?

And is it important to have members of the senior management team in each of the groupings of 10 or 12 teachers?

Well, I think in terms of grouping, we originally assumed that in primary schools, early years teachers should be in one group and maybe upper primary should be in another group.

And certainly in secondary schools, we assumed that the maths teachers should be in one group, the science teachers in another, history in another.

And it turns out that that was actually not at all necessary.

And we should realise that because what we're really focusing on here is pedagogy.

And so therefore, what we've discovered is that when you have a diverse group in secondary schools that brings in maths teachers and history teachers and English teachers and science teachers, and art teachers and music teachers, is that provided they come to the meetings with this principle of openness, that they actually understand that you might need to do some work to make somebody else's technique work for you, then teachers find a really quite powerful cross-fertilisation.

So originally, we did think that there would need to be a specialism for each group.

But now, we've seen that that's not appropriate, in fact, because the point is there is no curriculum for this group.

Each teacher comes along with what it is that they have decided they want to work on to improve their practice.

And the job of everybody else in the group is to be a supportive colleague in terms of helping the individual teacher do what that teacher has decided for themselves they want to do.

The issue of senior management is quite interesting.

In many primary schools in particular, we find that heads don't like being excluded.

I mean, in small primary schools, the heads are just like everybody else involved in everything.

And the reason we call these groupings teacher learning communities is that they are designed to be by teachers for teachers.

And so what we say is that if you are trying to change your practice, if you have a group of students that you're teaching, and you're trying to change your practice, then you can be a full member of this group, because it's about people who are trying to change their practice.

But if you no longer are teaching, then it's not about you.

You cannot be a full member of this group, because it is not about you.

And I think that's particularly important because once one stops teaching, and you could argue once one stops teaching full time, but certainly once one stops teaching, one actually, I think, loses all credibility as a teacher.

As soon as you stop teaching, you forget how difficult it is.

And therefore, we think that the most effective leaders of these groups are not people who are seen as particularly skilled or effective teachers.

In fact, we have seen teachers in their second year of teaching being very effective leaders of these groups, because they know they don't have the answers.

They don't actually seek to tell others what to do.

They seek to create a supportive group in which the learner gets support in doing what each teacher gets support in doing what they want to do.

So this is a group in which there are no experts, just people who have made a professional commitment to improve their practice.

And they get support from like-minded peers.

It really is a very radical model.

And people say, how do you know they're going to be doing the right thing?

I don't.

But I actually trust that given the time to think about this stuff, most teachers are professional enough to spend the time productively.

Now we're just coming to the end of the show, Dylan.

I'm just wondering about when you're talking about teacher learning communities, is there still a place for a practitioner or action research around formative assessment, either at school level or actually feeding into the work of yourself and other academics?

I know you've got a long history of working with teachers and that kind of idea of teacher agency and actually the teachers taking on the ideas and developing new ideas themselves.

But I just wonder how you feel about the theories of formative assessment.

Are they fairly settled and uncontested?

Or do you think what's going on in teacher learning communities needs to continue to be researched to make the theory an iterative one?

I mean, I'm absolutely clear that we don't know a whole range of things about formative assessment.

For example, we don't know how important subject-specific knowledge is to doing formative assessment effectively.

We still don't have an agreed definition of what we mean by formative assessment.

So I think there's a huge amount of work to be done.

With regard to the role of action research, I actually don't think that teachers should be required to engage in action research if by action research, the research activities or inquiries that are designed in such a way so as to produce generalizable findings.

I think if teachers want to do that, they should, of course, be encouraged to do that.

But for me, the essence of research is that it transcends the context of data collection.

And therefore, if teachers want to do that, then, of course, they should be supported.

But I don't think teachers should be required to do action research.

I think teachers should be required to research and improve their own practice.

I don't want to require teachers to write that up.

I don't want them to have to produce publications.

But I do think that we need to be a little bit smarter about distinguishing between professional development for teachers, which need not satisfy the traditional talents of research.

And we need to be clear that that's not the same thing as action research that might lead to a publication in a journal, for example.

So I think both are useful.

I think the teacher-academic research can be very powerful here.

Ultimately, what I see Paul Black and myself doing is not leading practice at all.

One of the things we've been doing is to follow behind the best practitioners and to make sense of what they do.

And then because we have some understanding of the theory behind this, we can broker this.

Let me give you an example.

Okay, Paul Black and I have been saying that comments are more useful than scores or grades for a long time.

So many teachers have taken to comment-only marking.

One teacher, Sharla Kerrigan, a teacher at a school in London, Enfield, had been doing this for many years, but she still wasn't happy with how much attention the students were giving to the comments.

So she came up with an interesting variation of comment-only marking.

What she did was with a, I think it was an S4 class, they'd done an essay on Shakespeare.

And she'd written the comments on strips of paper rather than on their jotters.

And then each group of four students got back their four essays and the four strips of paper, and the group's task was to match the comments to the essays.

Now, I think that's a wonderful variation.

My guess is that she came up with that intuitively.

I can take that idea and see actually one of the reasons I think it's so effective is because what she's done is to slow down the ego involvement.

So when a student picks up one of these comments, they have no idea whose comment it is.

So they're reading it to find out what it says, not what it says about them, because they don't know whose comment it is.

So it seems to me that the work of researchers is different from that of teachers.

The job of teachers is to teach and to innovate.

And the job of researchers like me and like Paul is to make sense of what these best teachers do and then figure out why this is so smart.

And then maybe to suggest that some bits of what this teacher is doing are not necessary, because these are just add-ons that are just that teacher's idiosyncrasies, but these other aspects of what that teacher is doing is essential.

So that seems to me a really productive relationship between researchers and teachers.

Teachers innovate and researchers make sense of this to make it available to a wider audience.

Yes, that's fascinating.

Dylan, just before we wrap up, I wonder if you've got any reflective questions for our listeners or maybe some key links or indeed a bit of information on current projects.

I know that there's an ESRC project on learning how to learn, which seems to tie in with formative assessment.

I try to make all my work available in an open way on my website dylanwilliam.org.

And so that would be the point I would start to suggest to most people.

The learning how to learn stuff, it actually got very messy in the end.

Learning how to learn sounds like a great idea, but it turns out to be extraordinarily difficult to make it work in a research context.

You know, there are people talking about these 21st century skills, but they turn out to be much less generalizable than one would hope.

So at the moment, I don't have any clear advice about learning how to learn things.

And the other thing that I would say that I think is working very well is the building learning power work that Guy Claxton is doing.

They are focusing on resilience and reciprocity and these other kinds of important skills.

And Guy's work is clearly getting a lot of traction and schools are being able to integrate this into their work.

And they're improving student outcomes as well as increasing the depth of the learning and preparing students to be learners in the 21st century.

It's been a real privilege to have you on the show, Dylan.

So thank you very much for giving up your valuable time and I hope the rest of your time in Scotland this week is pleasant.

If you have listened to the live stream, you'll be interested to know that you can share the archived audio with colleagues.

That will be at edutalk.cc in a few days.

If you have any further comments for Dylan, and I know that a couple of people on Twitter have been listening live, but we don't have any questions, unfortunately.

But if you'd like to continue with any of the themes of conversation, then you can email edutalk or put something up using edutalk hashtag.

And indeed, as I understand it, Dylan, you're on Twitter and it's just at Dylan William.

So thanks again, Dylan, and everybody who's listening, take care and see you next time.

Thanks.