Yet another planning meeting. This time, you’re choosing next year’s professional development, either:

a) Instructional coaching
b) Teacher learning communities
c) Lesson study

Your team is the epitome of rationality, so you begin by thinking about the evidence. Sean describes the research review showing the effectiveness of instructional coaching (Kraft et al., 2020). Meera points out that Kraft and co didn’t actually define instructional coaching. She finds the EEF’s trial of formative assessment teacher learning communities more convincing (Speckesser et al., 2018), particularly since it was conducted in English schools. Ellie chips in to make the case for lesson study. That’s harder, since an EEF trial showed no impact on student learning (Murphy et al., 2017). But Ellie suggests the school could learn from that trial and design something better.

You hide your sigh. You asked for evidence – you got belief. For Ellie, who has been at the school a long time, good professional development must be gradual, consensual and teacher-led… must be lesson study or teacher learning communities. For Sean, who introduced coaching in his last school, professional development must improve teaching quickly… must be instructional coaching. Your team are prisoners of their narrow experience. So are you. You describe your experience, you listen respectfully to one another – but you struggle to understand one another’s understanding, or to truly learn from it.

If only someone could present compelling evidence, answering this question once and for all.

Seeking compelling evidence, answering this question once and for all

In our recent systematic review of teacher professional development, we set out to tackle this question: which form of professional development is most effective?

We broke each professional development programme down to its building blocks: the mechanisms used to help teachers change. This took us beyond vague statements, like teachers ‘attended training’ and ‘collaborated.’ Instead, we pinpointed what happened in training: modelling, planning, practice? And we looked at how they collaborated: giving feedback, planning lessons, offering practical advice?

Forms like ‘instructional coaching’ and ‘lesson study’ sound carefully thought through, but are also pretty vague. The individual coach probably knows what they’re doing. But when I say “instructional coaching,” one listener hears precise feedback helping me improve, while another hears having your next move dictated to you.

But we can get beyond this vagueness. We can break each form of professional development (coaching, lesson study, etc…) into its building blocks: the mechanisms. We can describe what happens when someone is coached, or they meet, or they study.

For example, researchers writing about lesson study typically describe a group of teachers working together to plan, observe and reflect upon a lesson. This incorporates three of our fourteen mechanisms/building blocks:

  • Feedback
  • Action planning
  • Practical social support

We characterised instructional coaching as incorporating goal setting and feedback, instruction or modelling, and practice inside or outside the classroom. Loosely, teacher learning communities just include meeting with peers to offer practical support. We adopted a stronger definition, which also included goal setting and action planning. (Sceptical about these definitions? You can find the full explanation nestled on pages 22-24 of the 196-page report.)

These descriptions are limited. You couldn’t design an entire professional development programme based on them. But by making them precise and narrow, we can be clear about which programmes count as one of these forms – and thus test their impact.

Which is best?

Before I show you, place your bets: which form is most effective?

My money was on instructional coaching. I’d run teacher learning communities and seen faster and deeper impact when we introduced instructional coaching.

The graph below shows the answer…

Each vertical line is a confidence interval: we know that the impact of that form of professional development lies somewhere on that line. Each black square is a point estimate: that’s the middle of the confidence interval. n is the number of studies in each category (16 for instructional coaching). k is the number of effect sizes in each category (35 for instructional coaching). k > n because one study may report two experiments, two treatment groups and so on.

No answer.

Well, two things stand out:

First, all three forms improve teaching. On average, lesson study, instructional coaching and teacher learning communities all lead to increased student learning.

But second, no one form is more effective than another. (The confidence intervals overlap – and you can see that the point estimates are fairly close to one another.)

Flash back to the meeting we began the post with. You cried out for evidence showing which form of professional development we should choose. You read this far into this post… And this is all we could offer.

Well, not quite…

If not forms, what does make a difference?

We did manage to come up with a real, useful, usable answer.

We’d found that the more of these building block behaviour change mechanisms a teacher development programme used, the greater the impact on student learning.

Now let’s look again at these three forms. The graphs below show what happens as you add more mechanisms (moving to the right) to those intrinsic to each form…

Each form of professional development has a greater impact, the more behaviour change mechanisms we add. Let’s say I’m using lesson study – so we definitely have feedback, action planning and practical social support. Now I add goal-setting, modelling and practice. I just increased the chance the programme works. I’m a generous person, so I add in praise and prompts. Even greater chance the programme works…

First, this is more evidence that the underlying finding – behaviour change mechanisms matter – holds. No matter which form you start with, more mechanisms help. Second…

It ain’t what you do – it’s the way that you do it

The phrase “It ain’t what you do, it’s the way that you do it” bothers me. Usually, it’s used to make a broad point: that there are no cast-iron answers; you have to choose wisely to meet your school’s needs, and implement effectively. But I worry about a potential inference: that there’s “No such thing as a wrong answer (in English/history/whatever…).” (Not true…)

Slightly against my better judgement though, I want to use that line here – but in a more precise sense. It ain’t what you do (coaching/lesson study/teacher learning communities) it’s the way that you do it (the mechanisms of change you use). It doesn’t matter what you call your professional development approach. It may not even matter hugely whether you choose instructional coaching or lesson study. But it does matter – very much – which behaviour change mechanisms you use as part of it.

If you adopt lesson study – but just pop into the lesson and summarise what you saw afterwards – not much will change. If your discussion includes several mechanisms – feedback, praise, examining models of good practice and planning future lessons, for example – it’s much more likely to have an impact.

This resolved one of my biggest worries about our approach. I worried that instructional coaching and well-run teacher learning communities are – underneath it all – pretty similar. Teachers set goals, get help, plan, act, get feedback and refine. But our approach suggests this is OK. They look similar because – despite their differences – they both employ multiple behaviour change mechanisms.

(In case you’re wondering, my money is still on instructional coaching, due to its frequency, personalisation and your accountability for acting on your goals. More evidence that people are impervious to evidence I guess.)

Conclusions

1. Arguing about whether instructional coaching or teacher learning communities is entertaining. But it’s a waste of time. No one form of professional development is a magic bullet.
2. What makes a difference isn’t the form – the core two or three mechanisms you use. It’s the way you use multiple mechanisms – and the way they cohere – to make change more likely.
3. Improving teaching is a work of behaviour change.

If you liked this, you may enjoy

The systematic review of professional development (with Sam Sims, Alison O’Mara-Eves, Sarah Cottingham and other colleagues) on which this is based – and the EEF guidance report derived from it.

This post on mechanisms of behaviour change in professional development: what they are, how we defined them, what they do.

This post on the three questions we need to answer if we want professional development to work.

This post describing how instructional coaching works.

References

Kraft, M., Blazar, D., Hogan, D. (2018). The Effect of Teacher Coaching on Instruction and Achievement: A Meta-Analysis of the Causal Evidence. Review of Educational Research, 88(4), pp.547-588.

Murphy, R., Weinhardt, F., Wyness, G. and Rolfe, H. (2017). Lesson Study: Evaluation report and executive summary. Education Endowment Foundation.

Speckesser, S., Runge, J., Foliano, F., Bursnall, M., Hudson-Sharp, N., Rolfe, H., and Anders, J. (2018). Embedding Formative Assessment: Evaluation report and executive summary. Education Endowment Fund.