After being in hundreds of schools and thousands and thousands of lessons, my experience would lend my support to the hashtag #nobestwayoverall. However, experience is nothing like enough. I have only been in a very small proportion of all the country’s schools and in a minute proportion of the lessons taught. No matter how important one’s experience seems, it is bound to be limited and mine is no different. Experience shapes you, but it mustn’t define your belief about the importance of something in the wider scheme of things. Belief must be tempered by evidence and as I’ll show here, there is little evidence for the widespread adoption of one particular method. Improvements in England’s schools have been due to the use by teachers of a wide teaching toolkit.

I’ll stick to meta-analyses, rather than focussing on the research of individuals, as the meta-analyses provide outcome data from many other studies as to the importance over time of methods of teaching. (a meta-analysis is the statistical procedure for combining data from multiple studies). I’ll also stick to improvements (or otherwise) over time in a large number of schools, rather than improvements in particular schools. Again, a focus on individual schools  would skew and distort overall evidence, as the sample sizes are far too small. Just because a method of teaching has had success, or may be currently enjoying success, in a school, or schools, does not mean it is a method suitable for all. Will it scale up? Are there enough disciples/adherents/potential converts to that style of teaching to allow the idea to spread far? Are there other reasons for the success of the school besides the teaching method (e.g. the quality of leadership and leaders’ – especially the headteacher’s – ability to recruit and align staff to a vision may be far more important). I’m thinking here especially of Steiner schools, Montessori Schools and more recently, a very small number of schools who appear to be achieving success via the introduction of so-called ‘traditional’, (‘trad’) teaching methods. All of these can be successful – there are outstanding Montessori and Steiner schools and I have no reason to doubt that some of the ‘trad’ schools will achieve this mark. Alignment to a vision is a powerful factor in the progress of a school and talented and strong leaders provide the basis for that alignment. Excellence can thus be achieved in many ways.

Evidence from Meta Analyses.

The two biggest, recent, meta-anlyses have been by the Sutton Trust, via the Education Endowment Foundation (EEF study, or EEF) and by John Hattie (Hattie), who has been Professor of Education and Director of the Melbourne Education Research Institute at the University of Melbourne, Australia, since March 2011. Both are widely respected in educational circles and despite some criticism of methodology, often acknowledged by the authors, leading to refinement, the both remain a benchmark for research into the effectiveness of methods.

EEF’s toolkit can be found here:

The toolkit is a summary of educational research on teaching 5-16 year old pupils. There is an Early Years toolkit as well and there are EEF projects on a selection of themes chosen in collaboration with teachers. It is focussed upon what approaches will work best with disadvantaged children, but disadvantaged pupils make up a large proportion of all pupils in the country (29%. or 2m pupils) and thus the findings can be extrapolated, with confidence, to all pupils. I have selected out only the approaches that work at least moderately well, with an effect size of at least 5 months. Other approaches do add value, but not to the extent of these. The numbers following each toolkit strand are the number of months increase that the application of each strand can bring for a pupil:

Collaborative learning +5

Early Years intervention +5

Feedback +8

Homework (secondary) +5

Mastery learning +5

Meta-cognition and self-regulation +8

One-to-one tuition +5

Oral language interventions +5

Peer tutoring +5

Reading comprehension strategies +5

EEF does not cover all the methods covered in Hattie’s study, neither does it focus on individual teaching strategies. I accept that it may be possible for skilled teachers to introduce each toolkit strand into, say, Kagan/collaborative learning, discovery learning, or direct instruction, but inference would say that this would be difficult with so many types of approach. The overwhelming message from EEF is that many approaches work. Why would anyone ignore this research and not make space for a range of methods their teaching?

Hattie takes a different approach. It is a meta analysis of meta analyses and its scope is huge. His research is now based on nearly 1200 meta-analyses – up from the 800 when Visible Learning came out in 2009. A link to Hattie’s research can be found here:

Hattie Ranking of Effect Sizes


There are 195 elements to Hattie’s research and he stresses that ‘all work’ (at least the positive ones do) but some have more effect than others. An effect size of 0.4 is the average, so Hattie advises educators to investigate elements of teaching that have a greater effect than 0.4. It is clear that some approaches have more success than others, but the differences between those that have the most effect are not big enough to say that one approach is the most important. They can all be important. It is down to individual schools to determine what is important in their setting i.e. context drives importance in one school. In addition, you simply don’t know which approach, or technique will work for a particular child. There are many elements that are not found in the EEF study. For instance, Piagetian programmes at +1.28 and direct instruction  at +0.59. Both are important however and in the right context, can be hugely beneficial as approaches for accelerating progress. Again, why would anyone ignore this research and not make space for both in their teaching, as well as paying attention to the other areas which Hattie’s research shows can bring benefits for pupils?

Both EEF and Hattie lend great support to #nobestwayoverall , as they are outcome measures. Research such as studies in cognitive science provide good pointers for teacher practice, but it is difficult to find studies that analyse outcome measures for cognitive science in terms of it’s effectiveness in overall school improvement. I’d be grateful to anyone who can point me to such studies. That brings me to another support for #nobestwayoverall. 

Schools in England have improved.

I’ve focussed on England, rather than the rest of the UK, as there is an enormously large set of data over time that supports the assertion that schools have improved and that is the changes in grades that schools have achieved since Ofsted was established, under John Major, through an Act of Parliament in 1992. Pre 1992, national data on the quality of schools is very sketchy and not reliable, but anecdotal and achieve evidence shows great variation in school quality and a significant proportion of schools that today, would be considered appalling, never mind inadequate. It is especially difficult to gauge how many schools were excellent. Compared to present-day measures, I suspect there were far less than now and the numbers may be small.

In  2009, after most schools had been inspected under the proportionate regime that was introduced in 2005/6 (good and outstanding schools inspected every 6 years, satisfactory schools every 3), 66% of schools in England were good, or better. Prior to 2005, figures are very hard to come by (I failed miserably, I’m afraid!) and the inspection grade scale was different (1-7), so not easily comparable. Those who would wish to know more of Ofsted’s history would do well to read the RISE Trust’s report of 2012 ‘Twenty years inspecting English schools – Ofsted 1992-2012’. I know I’m a nerd, but I found this to be a truly fascinating document to read! http://risetrust.org.uk/pdfs/Review_Ofsted.pdf

In HMCI’s report of 2016, 91% of primary schools and 78% of secondary schools were judged to be good, or better. A significant improvement over a period of time where the Ofsted framework has got tougher. If those schools were inspected against the 2006, or 2009 frameworks, it is highly likely that even more would be good, or better. If schools haven’t improved, how can this be? The only explanation would be that Ofsted have been constantly getting judgements wrong, or examining the wrong things. However, the framework, though not as full as I’d like it to be (workload and well-being come to mind), it is admirably wide and allows comparison over time. Improvements in the percentages of schools being good, or better, shows that achools overall have improved

Further evidence of school improvement is that results at KS2 and KS4 have also improved and add extra support to schools having improved – and improved significantly – during the last few decades.

The norm in teaching is to adopt a wide range of techniques to suit pupil need and context. #nobestwayoverall

During this time, 1992-present, teachers in England, via the National Strategy work and beyond, have adopted a wide range of teaching techniques, applying them, I feel with genius, to the learning of classes and especially of individual pupils. This approach has worked, as schools overall have clearly improved. Within this there are schools who have adopted and developed particular methods of teaching and learning too. Some are well documented, others nascent and awaiting evaluation. As Hattie said, all elements of teaching and learning can work; they just have to be applied in the correct way, by good teachers and good leaders of schools. Some schools fail when approaches are not well applied and this is usually down to poor leadership – here, a change to a different approach can certainly pay dividends.

If schools have a track record of success, why not build on that, instead of abandoning an overall successful approach in favour of teaching in a particular way? The response to that, from a school which has gone down that route could easily be ‘look at what our pupils are doing now’, but that would be to fall into the trap I exposed in my first paragraph: one’s own experience is not enough to define one’s belief. Evidence is needed that your belief is likely to be correct. However, when a school is well-led and staff, parents and pupils are well-aligned to a vision, that’s a different kettle of fish and a singular approach can be very successful (e.g. Steiner, Montessori).

A wide range of teaching and learning styles has produced improvements in England’s schools. Marry that with great leaders, who trust teachers to take control of their own personal professional development (PPD) and thus chart their improvement in ways best suited to themselves and IMO, you have a powerful cocktail for improvement. To do this, teachers have to see other ways of teaching in action and be able to reflect on what they see with others and the teacher whose lesson they are visiting – but that’s another story. That’s Talk for Teaching!