Improving the learning of teachers

Lee Elliot Major lists ten tips from research

One of education’s great ironies is that we invest huge efforts to improve the learning of pupils yet neglect the learning of teachers themselves – the very people we entrust with our children’s development.

The ‘core business’ of teaching matters more than anything else within schools, particularly for pupils from disadvantaged backgrounds. A Sutton Trust review found that over a school year poorer pupils gain 1.5 years’ worth of learning with very effective teachers compared with 0.5 years with poorly performing teachers. In other words a great teacher can produce a whole year’s extra learning. That’s a big, life-transforming difference.

Teacher time

The problem for the 450,000 teachers in the UK is that they are not given much time to reflect and refine their craft. The latest Teaching and Learning International Survey (TALIS) by the OECD found that teachers in England have larger workloads than their peers in many of the world’s highest performing nations in education. But despite this, what is striking is the extra time devoted in these higher-performing nations for teachers to learn from other teachers through observation and feedback.

It is not just quantity but quality that is a problem. The uncomfortable truth for educators is that little has been done to evaluate the best approaches to teacher learning. It means that many well intentioned professional development programmes are likely to be ineffective.

Ten things teachers should know from the research

Some general lessons however have emerged from the studies across the world. Here are ten things all teachers I believe should know.

1. The best bets for teacher learning are those that involve teachers working alongside other teachers focusing on their practise in the classroom.

2. Reviewing a teacher should ideally combine the results of multiple approaches of assessment.

3. The strongest evidence supports the use of three main sources of feedback: teacher observations (particularly from fellow teachers or peers); surveys of students; and measures of student progress.

4. Observations and feedback for professional learning should be a separate process to appraisals or performance management.

5. Teachers should be properly trained as observers.

6. Teacher observers need to be extremely cautious about their assumptions of what great teaching looks like. In his blog ‘Classroom observation: it’s harder than you think’ Durham University’s Professor Rob Coe spells out the many traps observers fall into.

7. Teachers should adopt the same principles for teacher feedback as they do for providing feedback to pupils in the classroom. See for example the “professional learning cycle” used by schools in Australia. It mirrors so much of what we know about feedback loops for dialogue between teachers and children.

8. It is critical to create a culture of trust and collegiality among teachers if feedback is to genuinely impact on practice. It is a key point highlighted in reviews undertaken by New Zealand academic Helen Timperley (referenced extensively by John Hattie, among others).

9. At the same time there must also be genuine challenge for teachers if they are to change behaviour, something that can be provided by school heads or external peers.

10. Last but not least, any professional learning programme needs to be rooted from start to finish in student outcomes. The yardstick for success is improved progress of learners in the classroom.

Sutton Gates summit

The research underpinning these ten points is summarised in the Sutton Trust’s report published this week. But useful as they are, they do not offer the practical strategies that school leaders and teachers have found to be successful.

More will be revealed by evaluations from the Education Endowment Foundation. One study, for example, is assessing the impact of Lesson study, a collaborative peer observation programme currently much in vogue in English schools. Another study will assess whether teachers trained as observers also improve their own classroom practice.

But with such little research available, what wisdom can we gather from teachers and school leaders? This is exactly the question at the heart of an international summit the Sutton Trust is organising in partnership with the Bill & Melinda Gates Foundation.

To be held over two days in Washington DC next week, the summit will bring together 80 school leaders and teachers from a range of countries including the UK, US, Canada, Australia, Finland, Holland, Singapore and Hong Kong. The participants will hear the latest research. But they will also bring with them their practical tools and strategies for observation and feedback that have improved teachers around the world. The aim will be to create a practical guide for the effective professional learning of teachers – currently so conspicuous for its absence in our schools.  I look forward to reporting on what we find.

2017-12-07T15:34:47+00:00 October 31st, 2014|Categories: Blog|


  1. Ben October 31, 2014 at 9:07 am

    Lee, interesting research you highlight today. Among other points, as a State Secondary School maths teacher, I am fascinated by the comments on setting. The Durham report’s comments are detailed enough to cover intra-class as well as inter-class setting; but can they go further: to inter-school setting? How do they fit in to the Grammar school debate? And, further, to Universities? To the creation and sustaining of elite establishments around the world? And to the field of employment? For example, to the best City firms? or best research groups? Or the best “do” tanks? I am, as you will judge, cautious; particularly with respect to mathematics (where there is some body of, somewhat mathematical, research). I remain wary of the effects on teacher workload (particularly in mathematics) of mixed sets; particularly those containing pupils who have special educational needs or who merely still struggle with multiplication and hence need very different work to understand, say, set theory addressing different types of infinity (new Key Stage 3 curriculum) that may refer to prime numbers. I am also intrigued as to how comments on setting apply/change as pupils grow older (progress from EYFS to A-Level). A-Level lessons necessarly move fast. Having a pupil in the same classroom with an F-grade at GCSE mathematics would, one might judge, require setting very separate work. “Setting” different work; is that another way of writing differentiating (differentiating; is that another way of writing setting?)? Why is it that mathematicians always see differentiation for, and integration of, pupils as part of a different form of calculus? I suspect that different calculus, often, relates to behaviour management. Thus speaks the grown-up version of the “bright” boy who gave up GCSE German after having had a lovely young man placed next to him who jabbed his hand (no, it wasn’t Nigel Farage) and who was told by another well-built fellow pupil he would be “beaten up” if he got 100% again in a maths test. Thus also speaks the teacher who has seen the behaviour of “sink” year 8 classes seemingly transformed by mixing them with pupils who’ve achieved better in tests. Speaking of tests; why do we test at all if setting is of no value? I was sure it was something to do with distant employers wanting a sense of ability. Why would they want that? There must be something of value in it? Yes, you are right Lee, education is complex, every child is unique and (ideally) every child would be have work set to develop just them (sometimes to do alone, sometimes in teams, sometimes in chaotic self-forming groups, sometimes to help others, sometimes to help themselves, some times to…). I do not envy you handling the remarks from those with PhDs and Professorships with respect to maths teaching at different ages. Trinity College, Cambridge and a tutorial of five students. Mixed setting of those aged 19?!

  2. John Bald November 21, 2014 at 8:51 am

    The Sutton Trust needs to carry out research into the effects of setting in current contexts, notably ARK and Harris Academies. The most recent UK data dates from the 1990s, and does not cover subjects beyond English maths and science. There is no data whatever on the effects of setting ln learning languages.

Leave A Comment