The culture of observations at Michaela is truly phenomenal. It provides a structure conducive to genuine improvement and honest reflection. Observations at Michaela are frequent, low-stakes and random. Since starting in September (around 24 teaching weeks), I have been observed around 70 times by other Michaela teachers. That’s right – 70! Observers are typically in my lesson for around 10 minutes, though I have had many longer observations. Every time I am observed, I am pinged an e-mail with ‘strengths’ and ‘improvements’. Equally, I probably observe around 2-3 teachers per week, and ping them immediate feedback. Observations and feedback culture are not dictated by hierarchy – every teacher supports every teacher at Michaela. In this post, I suggest three reasons why this observation culture is more powerful than a less-frequent, high-stakes, graded observation system.
Reason 1: High-frequency observations allow recurring themes to be unveiled.
In a model of observations that occurs three times per year, the major risk is that anomalous successes and unfortunate errors are stamped onto a teacher and transformed into a medium-term plan of action. The problem is, that anomalies are mistaken to be the norm, and this risks missing genuine areas of improvement. In the same way that we would evaluate summative assessments as being poorly designed if they sampled too small a domain for a valid inference to be made about a longer sequence of teaching, so too is a low-frequency observation model flawed.
Where observations are more frequent, strengths and areas of improvement can be identified when similar feedback is given several times. This means that if you fumble your questioning once in your observations, where normally this does not occur, this feedback can be taken as an instance, with no sense of dread that this will be deemed reflective of your teaching. Your excellent questioning will be noticed in other lessons later in the week. On the other hand, if you repeatedly receive feedback: ‘Your explanations are quite lengthy – break it up with questions or pupil practice’ several times – you can be sure that this is something you need to work on.
Reason 2: Low-stakes observations create a culture of learning from colleagues.
In schools where observations are used as a high-stakes way of evaluating teaching, there can be a culture of fear around observations. It discourages teachers from visiting each other’s classrooms since observations become associated with a grade and/or with judgement.
In a low-stakes observation model, the natural culture to emerge is one of support, improvement, feedback, progress and learning. Teachers at Michaela love other teachers popping into their lessons. In fact, members of my department have sent department-wide emails asking why so few teachers have visited them this half-term?! This evidently reflects a willingness to receive feedback to improve and to be reassured of what is being done well.
Reason 3: The randomness of observations makes them stress-free and useful.
I remember spending HOURS planning my first ever graded observation lesson in my NQT year. It was quite stressful, as it was one of three observations that would label me with one of ‘Requires Improvement’, ‘Good’ or ‘Outstanding’ for the remainder of the Autumn term and most of the Spring term. The unintended consequences of this planned observation are: I probably neglected the planning of other lessons that week; the observation would not be representative of my typical teaching and planning; the nerves. Of course, random, graded, high-stakes observations would be worse; my critique of planned observations is not independent of the grades and low-frequency.
The randomness of Michaela observations means I have no incentive to prepare disproportionately for some lessons compared to others. This doesn’t mean that I never request colleagues to observe me at specific times; sometimes I want a colleague’s feedback when attempting a particularly complex explanation or trying something new. But, I am now so used to teachers (or visitors!) walking into my classroom, that I sometimes forget they are there after they have walked in.
The bottom line is that eliminating planned, high-stakes, graded observations is easy. The evidence is compelling; the lack of reliability of grades suggests they are flawed. Evaluating teaching and learning can be done easily from frequent observations and feedback – teachers get better quicker.
The culture of learning from each other that emerges is the most wonderful consequence. Some of the best conversations I’ve had about teaching and learning occurred after an observation. There is vast expertise to be found in the walls of every classroom – it should be every school’s priority to ensure the structures are in place to ensure that all teachers feel comfortable to walk in and learn from each other. Equally, every teacher, no matter how excellent, should feel encouraged to open their classroom to welcome feedback from all colleagues. And the best practitioners do indeed crave the feedback.
Are these reasons not sufficient?!
Thank you to all the teachers who have spent their time observing and offering feedback to me. Two claps on the count of two…