Six Models Of Lesson Observation

Reading time: 4
Lesson Observation Outside Teaching


In 2010, Ross Morrison McGill founded @TeacherToolkit from a simple Twitter account through which he rapidly became the 'most followed teacher on social media in the UK'. In 2015, he was nominated as one of the '500 Most Influential People in Britain' by The Sunday...
Read more about @TeacherToolkit

What do OfSTED want to gain from lesson observations?

This is a summary of 6 observation models presented at an OfSTED seminar to discuss how lesson observations may help with future inspection development – curated to support reliability and validity – as OfSTED seek to strengthen their purpose.

What is reliable?

Over 25 years, I’ve observed thousands of lessons. The skills required to be able to do this effectively for the benefit of the teacher have been a constant learning curve – especially when watching outside of my specialism. This is clearly an interesting paper which has the potential to impact on future policy for all.

The critical question to ask when reading this research is, “is the purpose of observations to help teachers to improve (e.g. as a result of feedback) or do we want to use observations to provide information about quality of teaching?” As OfSTED state and rightly so, this research does not conclude Ofsted’s future model for observing lessons which needs further conversation, testing, consultation and decision-making.

What’s in it for OfSTED?

Her Majesty’s Chief Inspector, Amanda Spielman is “committed to constantly improving the validity of inspection so that our judgements of schools are the best reflection of the quality of education they can be. Scrutinising the reliability and validity of inspection methods is an important part of improving validity overall.”

Spielman recognises that grading one-off lessons or individual teachers is flawed, but says “where observation is used to inform judgements of school quality, it would be a mistake to pick up an off-the-shelf model from elsewhere and apply it wholesale. This does not mean that the lesson observation models presented did not have anything new to offer.”

Something to note, “how best to apply this new knowledge.”

Lessons from the past …

Ofsted has used lesson observation as part of the inspection process since its foundation in 1992 – a methodology almost 30 years old! “At one time (OfSTED, 2003), it was expected that 60% of inspection time in schools should be used to observe lessons.” Inspectors used observations to collect evidence on teacher behaviours and student interactions in a high-inference qualitative format.

How things have changed …

In 2005, a more proportionate inspection process introduced to a reduction in the time available for lesson observation. There was less expectation for a certain number of observations to be completed and lesson observation become less focused on subject-specific content.

And my friends, here lies the issue we still have today. I recall in 1997 and in 2002 when subject-specialists who genuinely work with teachers in their subject area to improve. Today, focus has shifted towards a generic set of attributes for teaching and learning across subjects to test quality (consistency) across the school workforce.

Fast forward to 2015 and inspectors no longer grade the quality of teaching and learning in individual lessons, nor judge the quality of teaching of individual teachers. Instead, inspectors now observe many lessons across the school to provide a reliable aggregate picture of teaching quality. Again, the scope of observation has moved away from individual practitioners and towards the school as a whole.

This, therefore supports the need for schools to seek teacher-compliance, when tested by a group of visitors over a one or two-day period.

I wrote in July 2017, if a teaching and learning policy is not statutory, why have one if it is used against you in an inspection?

Gathering evidence

One of the word ‘terms’ blighting our profession today. Teachers now back-track all the work they have done for an academic year, on the whim that an observer may speak, gather or view [required evidence]. “Inspectors are still expected to use a considerable amount of first-hand evidence to determine the quality of teaching and learning.” For example, observing pupils, talking to them about their work, scrutinising their work and assessing how well leaders are securing continual improvements in teaching. Moving more towards looking at students’ books to gather a picture of progress ‘over time’.

What if a subject teacher doesn’t use extensive pieces of writing in PE or drama?

Key research questions asked

  1. What can Ofsted learn from international best practice in the use of lesson observation for the evaluation and improvement of teaching quality?
  2. What changes should Ofsted consider in developing its inspection framework for 2019?

Using six models from

  1. Classroom Assessment Scoring System (CLASS)
  2. Framework for Teaching (FfT)
  3. International Comparative Analysis of Learning and Teaching (ICALT)
  4. International System for Teacher Observation and Feedback (ISTOF)
  5. Mathematical Quality of Instruction (MQI)
  6. Generic Dimensions of Teacher Quality.

Main findings

  • The six models provide a clear focus for those observing lessons.
  • The models all produce predominantly quantitative data – that’s measuring teachers folks!
  • The structure of the six models were typically informed by the research literature on the quality of teaching; evolved over years.
  • None of the models explicitly attempted to measure learning. It was generally agreed among the experts that learning is not something that can be directly observed, while the quality of teaching can.
  • Most models are similar in terms of the relationship between observation ratings and pupil attainment measures.
  • Despite the emphasis on quantitative measures, most of the models were still considered to be high inference (subjective versus low inference which captures observable facts or events, with minimal interpretation or subjectivity.)
  • The experts all accepted that their models could never be 100% reliable.
  • There was some debate over the use of video recording instead of live observations – although a remote observer could lose vital contextual information.
  • No agreement was reached on the ideal observation length of time or number of observations required.
  • Few subject-specific models currently exist beyond mathematics and language and literacy. I fear for any inspector who comes into my design technology lesson to tell me ‘how good’ the teaching is…
  • The cultural specificity around the models means that they are unlikely to work exactly as intended in different contexts.

As ever, this research supports my claim that coaching for all teachers, led by those working closely and frequently within the school itself as best placed to observe quality and drive teacher development. This is not something OfSTED will achieve unless it moves to a peer-led methodology.


To read the details of each of the six models, download the research document.

One thought on “Six Models Of Lesson Observation

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.