Are observations out of date?



The one thing you can always be sure of in teaching is change.  The utterance, ‘but that’s the way we’ve always done it,’ seems futile at the best of times. Certainly, such a viewpoint has often held back authentic impact on our children’s learning; teachers who hold on to their ‘ways’ without checking up on the effect are usually on a hiding to nothing. So, it is in this vein that I want to suggest that there could be great benefit from schools abandoning formally observing teachers in the way that many have done for so long.  I’ve no doubt this might cause a sharp intake of breath for some leaders, and I’m sure an excited squeal from teachers far and wide, but there is growing evidence to support change in observing teachers at work.

Having just been through an Ofsted inspection, and  being able to compare this with two other previous inspections, the emphasis on formally observing teachers has clearly changed and there is good reason for this, both in theoretical and practical terms. Education is emerging from being suffocated by a performance driven culture that squeezed the breath out of authentic pedagogy for learning. Great emphasis was put on data and teaching to evidence school improvement, while learning itself was assumed rather than confirmed. At last educators are beginning to put learning first and understand that when schools do this, everything else falls into place. This is why understanding the difference between learning  and performance culture is so important and often rarely discussed in schools.  I’ve written about performance in education before and one of my heroes, Chris Watkins spent many years unpicking this phenomenon in detail.

One might say that the culture of measuring performance has always been present through the system of summative testing; however, the obsession with judging schools and teachers through narrow performance measures grew sharply with the introduction of the national curriculum in 1988. This was fuelled by Thatcher’s dislike of teachers, her love of market forces, plus Kenneth Baker’s obsession with measuring and ordering anything that moved. I’ve no doubt education needed a kick; however, the publishing of league tables resulting from high stakes testing, and let’s not forget those terrible timed three part lessons, created a culture change that was more corporate and ‘sales driven’ than ever in the history of English education. Tony Blair of course came along and rather liked it all; ‘accountability’ was indeed  New Labour’s watch word, but accountability and learning often aren’t very good for each other at all. As Dylan Wiliam (2016) remarks, ‘it is highly unwise to use a teacher evaluation framework as a teacher improvement framework,’ and will most likely have the result that ‘we improve teachers in ways that do not benefit their students’.  Certainly, chasing teachers for levels as a way to ‘raise standards’ did just this. However, are we also doing something similar with formal observations of teachers?

If I look back at my career as a teacher and think carefully about how and why I improved from a bumbling, inexperienced NQT to where I am now, none of it was ever born out of the highly pressured observations that were tied to performance management or school evaluation. These were something just to get through and hope you were able to pull off on the day.  Yet where the real improvement and confidence came from was from my sincere wish to improve all the time, studying and understanding what really helped children learn well and watching and learning from other practising teachers. To my mind, this is the most effective ‘improvement framework’ that will lead to more effective teaching and better outcomes for pupils. It’s true that not all teachers are that reflective about their practice (although I haven’t met that many who aren’t) and not all teachers are interested in learning about pedagogy as such, but there are ways to encourage change here.

In the same way that children learn little when they feel under the microscope, teachers are the same. Teachers also usually learn more from each other rather than they do their leaders, some of whom may have been out of the classroom for a long while. Teaching certainly requires its knowledge base, the longevity of experience matters, but it is essentially a practical skill which requires practical transfer in the same way that learning to drive requires driving around with a driver, swapping and taking the wheel of course, but essentially learning alongside a practitioner. Who would ever feel confident learning to drive with an instructor who no longer drives?

However, school leaders certainly do need to see what’s going on in classrooms and understand the quality of teaching and learning in their schools – this is central to their role, but this is not the same as supporting teachers to improve (problems result from evaluation frameworks and improvement frameworks being transposed remember.)  So the question is: can leaders evaluate teaching effectively without formalised performance driven observations? The trouble is that a formal observation doesn’t always tell you that much about a teacher’s actual daily practice because the event is, by its nature, a performance, a presentation, a show and not an observation of ‘practice,’ otherwise defined in a dictionary as ‘the customary, habitual, or expected procedure or way of doing of something’. In fact, often teachers drop parts of their customary, day to day practice in observations because it may not be on what they perceive as the desired tick list for a lesson; however, the idiosyncratic way that a teacher works might be very effective indeed for their pupils’ learning because of the distinctive relationship they have with them. In other words, it’s unlikely that a school leader will really get what’s going on by sitting with a clip board and watching a single lesson. They will only really find out what the teacher did or did not do well that day; leaders might believe they can infer the daily practices of a teacher in that slot, but they just can’t.

It is in this way that wiser folk have turned to look at the progress in pupils’ learning over time using a ‘range of evidence of learning’ and this is a key phrase that is so important in any evaluation framework. If we know that watching a teacher perform doesn’t tell us that much, but a range of evidence of learning does, then we need to be clear about what evidence of learning really is. This is why formal observations may not be that helpful as part of an evaluation framework, but instead should morph into developmental peer observations and be shifted  into a school’s improvement framework.

In returning to the evidence of learning, this is still a kettle of fish, but things are looking up. Inspectors and leaders went through a very strange phase of looking for ‘rapid progress’ in a single lesson, so much so that children all over the country were rushed through content so quickly that it’s a wonder they learnt anything at all. Classrooms suddenly became quite unsettling places to be in. It surprised me that even some really experienced school leaders could not tell the difference  between teaching pace and pupils’ cognitive pace, which are quite different, not seeing that all that matters is that they are aligned, not that they are of any particular speed. Put bluntly, it really doesn’t matter if pupils are on the carpet for 20 minutes, what matters is whether they are engaged in the learning journey. Ofsted got this in the end and ensured inspectors stopped nit picking on style and timing, but instead looked at what they should always look out for: learning.

This brings me back to the utility of formalised observations to evaluate teaching and learning. It’s questionable if they really give leaders a true understanding of the quality of teaching and learning in a classroom; they can also skew teachers’ practice as they spend hours stressing over observations, and also the outcomes of observations, rather than aiming for consistent good practice over time. It seems a far better idea to have a situation where leaders are simply a common presence in classrooms through learning walks and by  being on the floor as often as they can, then their presence becomes normalised and teachers move away from inauthentic performances. By doing this, leaders also get a much more realistic idea of the day to day habitual practice of teachers and whether this is driving learning forward.

If the purpose of observations is to improve teaching and learning rather than evaluate it, then they shouldn’t be that formal and certainly shouldn’t be connected to performance management or pay increases. This avoids skewing teachers’ motivation because the nature of learning and performance is such that if you feel that you’re being judged on a performance you will behave differently than if you feel you are involved in learning and developing. The thing about performing is that improvement is not the motive whereas with learning it is!

Observations under an improvement framework should be driven by the teachers themselves, through reflective practice and professional dialogue. We have heard time and time again that teachers improve when they focus on their strengths, share this with other teachers and build their own practice. Of course a shared understanding of how children learn well and what a good lesson might look like should be constant features of discussions outside the classroom so that there is in effect an evolving school ‘success criteria’ by which teachers can assess themselves, but teachers will improve far quicker if they are encouraged to improve through peer and self assessment; the same is true with pupils.

In this way, leaders need to decide whether they are observing a teacher to help them improve, in which case they need to spend time with that teacher as a peer before and after the lesson discussing their practice and how they might improve. Alternatively, if leaders want to use observations as an evaluative tool, to assess teaching and learning, then they should not rely on these one off appearances in classrooms three times a year or so, but rather they should make themselves a common presence in classrooms and also rely on triangulating their assessment with the other forms of evidence of learning,such as pupils’ learning in books, conversations with pupils and of course data (that is quality assured through moderation and intimately connected to real learning).

At my school, I think we’ve done a really good job in separating our evaluation and improvement frameworks so that (as Dylan Wiliam suggests) we improve teachers in ways that do benefit their students rather than simply satisfying administrative or evaluative structures in school, but I’m not sure this is the case in that many schools. A clear distinction needs to made between appraisal and improvement when it comes to sitting down with that clip board. Sensible people stopped grading lessons last year, perhaps this year we should go a step further?

Ref. Wiliam, D. (2016) Leadership for Teacher Learning. Learning Sciences International.

Assessment: The journey so far…

traditional teaching

I often wonder why I’m so interested in assessment. Why I read about it in my holidays and feel perfectly happy tinkering and engineering my ideas about it all the time. It never feels like work. Well today, as I sit on a Greek island somewhere far away, I know why. It sounds too profound, too Hollywood, too idealistic, but to my mind getting assessment right will offer a great gift to our children. The gift will be to know themselves very well indeed; to look inside their own mind and understand themselves; to be deeply honest with themselves; to look towards quality and measure themselves against it; to work towards improvement because they know change is always possible; to see the change in themselves and to understand that improvement is always in their own hands. The right approach to assessment, throughout a child’s school career, has the potential to do this; it has the potential to make our kids strong people.

So, what about this year? How far have we travelled in this direction? Well, I think I can sum up this year as a time when we sought to close the gap, but this time, not just between what children know and need to learn, but at last, as an educational community we have thought about the gap between actual learning and all manner of ciphers for learning. In simple terms, this last two years have been about accepting fact that we’ve spent too much talking about units of measurement for learning rather than the learning itself. And in that mix, we often misunderstood assessment, or at least saw it only in one light, which was to act as a critical tool to hold ourselves and our teachers to account. I’ve no doubt that many teachers and leaders did recognise this and fought against the tide, but so many didn’t and perhaps still don’t.

Over the last two years, I’ve helped implement Learning Ladders, a new curriculum assessment system into our school, and supported some other schools to do the same. I consider assessment as a pedagogy, so that whatever we do we always remind ourselves that when we evaluate a child’s learning, we make sure that not only do we do something with that evaluation that benefits that child further, but we also do our best to involve the child in that process too, gradually drawing our pupils into evaluating their learning themselves and acting on that evaluation. I will always maintain that the best teachers seek to become useless to their pupils…eventually.

So now, where are we? What have we gained? What do we need to think about next? (I’m thinking both in terms of our assessment system and for assessment in general.)

The gains:

  • A large part of the educational community is taking charge, sharing ideas and practice. For example, look at the work of Beyond Levels and the #LearningFirst conferences. School leaders and teachers are coming together to share ideas and tease out the best ways forward for learners.
  • Assessment has moved away from being associated purely with data and tracking and is becoming increasingly associated with making an impact on pupils’ learning. Hurray!!
  • More teachers are being held accountable through the development of their pedagogical craft, with a view to improving learning (and learning behaviour) over time, rather than being held to account through straight forward numerical data, that may or may not accurately reflect a child’s journey towards academic progress. (I’ve just spent six months on secondment with a really challenging group of children, much of the progress they made was in their behaviour and attitude to learning; progress was certainly not always academic in some cases. Did I have any effect on that ‘progress over time’? Yes I did! Now they are well set up for next year when they will take off!
  • Senior leaders are able to lead teachers more effectively because gaps in pupils’ learning are easier to identify which results in more productive conversations about pupils’ progress. Our Learning Ladders system has a finely tuned gaps analysis tool so that when overviews of progress are looked at, conversations about why certain pupils are or aren’t making progress become very detailed about the aspects of learning in question. The result is really productive conversations about assessment, curriculum planning and pupil progress rather than those ones of the past where progress meetings were around levels or sub-levels and the details about pupils’ learning were not always foremost in people’s minds. We’ve discovered that being able to drill down to the granular curricular detail has meant that it’s much easier to pin point issues. Sometimes the issue might be teacher’s confidence in assessment; they’re just not sure about how to assess a certain aspect, it might be the first time in that year group and they’re finding their feet. Other times, it might be a teacher needs to refine and focus their planning a little more so they hit gaps in learning and at other times we might see that a child has been absent on the three times division and fractions were taught for example. This kind of depth of conversation just didn’t happen as easily with levels and for so many reasons.
  • School leaders can look at overviews of learning (which all leaders have to), but with Learning Ladders we have purposely not made inflexible bench marks or narrow progress thresholds for points within the year. Achievements in learning are noted on the system and accumulate through an algorithm into a score, but this is used as a measurement outline. This allows for the overview that school leaders need, because we have a traffic light score range based on a very general expectation of progress, but the  fact that it’s considered a range means that teachers focus on the learning rather than getting to a certain score;  plus, we have worked hard to make our assessment ethos mean that everyone understands the difference between ‘being seen to reach a level or a score’ and real progress in learning. These two were often confused under levels. Back then, moving up a level  assumed progress in learning, whereas now real progress in learning leads to an increase in the score. This might all seem like playing with words, but this is the whole impetus behind the idea of ‘learning first’… put the learning first and data will follow, but if you put the bench marks first, it might not.
  • Curriculum and assessment relate to each other in a cause (what do I want to learn) and effect (what did I learn) cycle rather than being loosely associated through summative assessment outcomes. This means that learning intentions are not merely derived from the national curriculum, but they are the curriculum. In the past there were two languages ‘curriculum’ and ‘assessment’ which meant that teachers had to translate the taught curriculum, into learning outcomes and then assessment judgments. Teachers no longer need to bridge the gaps between what is taught and assessment judgments because they are using the same language.
  • Teachers are more able to use assessment as a framework for planning because they are clearer on what children need to learn next and where there are gaps in children’s learning.
  • Teachers are able to access quality learning outcomes through shared learning moderation within our Learning Ladders group and soon these will be available to all on the system too. This means that the sloppy ‘best fit’ approach has been refined into a much sharper mastery approach for the detailed steps in learning. While I agree with many that the interim frameworks are far too demanding (that was my experience in Year 2 anyway), the Learning Ladders system means that the details required for a mastery curriculum to work well are exemplified. All assessment needs to be underpinned by shared images of quality and this should underline any decent assessment system.
  • After a year of everyone teaching the new curriculum, teachers are moving from using Learning Ladders as a ‘tick off tool’ to much more of a support for planning. Yes, we teach more than just the criteria on Learning Ladders because that is the basis for a broad and balanced curriculum, but that structure and mapping of the curriculum has been invaluable to support teachers mapping their way through all the changes.Teachers’ confidence in assessment and planning for it are now on the up!

Area of development:

  • The DfE interim frameworks don’t seem to reflect the key performance indicators considered appropriate by the rest of the education community. A lot of the guidance that goes with them is vague and open to many different interpretations. This has meant that teacher assessment is more difficult and less reliable as schools become more reactive to moderators requirements than authentic learning needs.Something isn’t right with those ITAFs! How many teachers have kicked themselves because they know that competent seven year old writers have had to be labeled ‘below expected’ because they didn’t do enough commas in a list or possessive apostrophes? This cannot be right.
  • 53% of pupils in the country reached expected in RWM the end of primary school. Really? Yes, expectations are higher, but pupils and teachers haven’t suddenly been knocked on the head so come on! Are we saying failure is a sign of success DfE? Schools need to plough ahead and make assessment work for their pupils; I know it’s hard – but we have to ignore this nonsense and follow our principles on assessment. We’re all in the same rocky boat of changing goal posts and incompetent management of national assessment from above, but we can still get on with doing what we know is right.
  • For some schools, assessment it still a vehicle for accountability much more than it is for learning. Leaders need to look at the progress over times in both hard and soft data and ensure this is aligned to authentic learning and not ‘ciphers for learning’. In other words, don’t set up a system that kids you into thinking all is well, when it isn’t!
  • Many schools still set children into ability groups and limit children’s learning through this approach. These schools need to trust learners and communities of learners and allow all pupils to reach their very highest potential; ability setting does not allow for this academically, socially or emotionally for pupils. Learning is not all about knowledge and skill acquisition.
  • Lastly, we have spent the past couple of years getting to grips with everything new, but we still need to move assessment more into the hands of pupils. Assessment is not complete unless it engages the learner into assessing themselves and moves them more and more towards independence. I think with Learning Ladders we have this in our sights. We have developed pupils’ overviews to summarise and see next steps, these have been very effective; next we need to refine these so they are easier for pupils to use regularly.  For me, this is the beauty of Learning Ladders, it is evolving to suit the needs of pupils, teachers….and leaders. This is the right away around, I promise you.
  • As always, I have to add that any assessment system can be used badly if the leaders running it don’t have sound principles on assessment; however, some systems encourage a certain approach that is modelled on the old levels system. No names here, but these should be avoided.

Final thoughts

I’m so optimistic that we can make assessment work for pupils in the UK, but we have to keep nudging the government our way and stand up for teachers in the classroom. Yes, we need to check teachers are doing the best by their pupils and then we need to check that school leaders are doing the best by their school communities, but as Mary Myatt put it so well, this must be through a culture of ‘High challenge and low threat’. The unwelcome consequences of a high threat culture in assessment mean that people then do things more out of fear rather than reasoned and deliberate action. High challenge, low threat always results in the best outcomes for pupils, teachers, leaders…and humans.