Instructional coaching seems to be the next big thing in education; it is the latest craze to ride a wave of enthusiasm across our schools. Faced with more and more schools ‘nailing their colours to this mast’, it feels almost brave to be holding back. But is it courageous and freethinking, or foolish and pig-headed, to go against the crowd on this?

In this blog, we’ll outline the thinking and rationale behind the flavour of coaching we have adopted within the Great Teaching Toolkit (GTT). But before addressing this in more depth, let us first try to define what we mean by instructional coaching. We may first note that instruction has quite different meanings in the UK and elsewhere. In the UK it means telling someone else what to do, and often has connotations of control and interference; in the US it simply means what in Britain would be teaching or pedagogy.

However, the confusion does not stop there. The same phrase, instructional coaching is used by different groups to refer to very different things. At one extreme, we have coaching guru Jim Knight (2019), whose version of instructional coaching emphasises teachers’ self-determination, autonomy, reflection and empowerment. At the other, Paul Bambrick-Santoyo (2018) uses the same words for an approach where a relatively directive coach identifies granular ‘action steps’ and prescribes a set of ‘coaching moves’. Others may use the same words to mean something between the two, or quite different from both—confusing, to say the least!

Nevertheless, there is good evidence to support the use of coaching as an approach to teacher development that leads to improved student outcomes. A meta-analysis by Kraft et al. (2018) found an overall positive effect of teacher coaching programmes—an effect that potentially out-performs other forms of teacher development or school-based interventions. This is a good study that presents solid evidence of effectiveness. An evidence-based practitioner would certainly want to look at coaching seriously. So, why are we not riding this wave?

Research-practice gap

Kraft et al. (2018) found that larger programmes, and those with non-volunteer recruits, have smaller effects, and pointed to the challenges for recruiting effective coaches. Indeed, they cautioned against seeing coaching as a general solution, “It may be that coaching is best utilized as a targeted program with a small corps of expert coaches working with willing participants and committed schools rather than as a district-wide PD program” (p. 574).

The researchers were unable to say which approaches to coaching are most effective, though there is some evidence that the expertise of the coach is a key driver (Blazar & Kraft, 2015). There is also some evidence that more effective coaches may also be more effective teachers (Blazar et al., 2022; Goldhaber et al., 2020).

The coaching models that have been evaluated generally depend on a supply of expert coaches, with experience and training, each working with a relatively small number of teachers on a one-to-one basis, over an extended period. One study (D. S. Knight, 2012) estimated the cost per coached teacher as up to $5,000—there are not many schools whose CPD budget can run to that!

The models that are being widely adopted in schools are quite different. In these ‘school-led coaching’ approaches, the coaches are teachers within the school, often senior staff. Leaving aside the question of whether those school leaders are also the most effective teachers or coaches, there are two obvious problems with this. One is the astronomical cost. It may be common in schools not to see people’s time as a direct cost, but of course it is. The other is the displacement. If we take all our best teachers out of the classroom to become coaches, any impact they have on the effectiveness of the coached teachers has to be offset by the loss to their own students’ learning.

When we consider all that, instructional coaching does not seem like such a strong bet.

A scalable, cost-effective approach to ‘coaching’

One of the design principles behind the Great Teaching Toolkit is that it has to be scalable: something that any school can do, that generates maximum impact for minimal cost and time. There is good evidence that one-to-one coaching by an expert is one of the most effective ways to improve teaching—but this ‘Rolls-Royce’ model is not a scalable approach. If we try to unpick the mechanisms by which coaching supports improvement, we may be able to adapt it to create something that can be of wider benefit to every teacher across a school.

Part of the attraction of coaching is its stark contrast with the ‘inspire and forget’ training sessions (generic topic, same input for all, one-off event, conceptual rather than practice focus) that are often the mainstay of school-based CPD. Obvious differences are that coaching focuses on practice and on solving problems that are salient in that teacher’s classroom. Coaching is sensitive to the individual context. The personal coaching relationship is both empowering and motivating. An expert coach can diagnose, structure activities and respond—just as an expert teacher does. When done well, coaching also mobilises four components that we know are important for all learning (but are hard to fit with the standard CPD approach): feedback, modelling, reflection, and deliberate practice.

Our challenge was to come up with an approach that retains all those elements but is feasible and efficient. We can still call it coaching because it contains all the same active ingredients—and because the definition of coaching is so loose and contested. Our Great Teaching Teams model uses reciprocal and group-based: teams of four to six teachers who coach each other, focusing explicitly on their classroom practice and their individual challenges.

The social aspect of the team motivates and encourages. We deliberately build expertise within the team with tailored courses and structured reflection. We provide feedback that identifies excellence, highlights progress, and grounds everything in a reality-check. We use video to make observation efficient, practical and powerful, as well as to share models of excellence. We invoke the principles of deliberate practice (Deans for Impact, 2016), including challenging and specific goals, ‘decompositions’ and ‘approximations’ of practice, and feedback and developing mental models.

Another key design principle (perhaps our secret weapon and a unique feature of the Great Teaching Toolkit) is that we systematically test and learn. We know that the way we configure, structure and advise gets interpreted and adapted when schools take it on, often turning it into something unrecognisable. We are not affronted or surprised by these adaptations; instead we work with teachers to understand them, to pave the paths they choose to walk, or, if necessary, to refine the choice architecture to better align practice and theory. Crucially, we build evaluation in at every stage, so that over time, with the collaboration of teachers and school leaders, we can learn which inputs give the best output.

We are confident the coaching model we ultimately converge towards will be scalable, efficient and effective. But we know it will not be the same as our current, evidence-based, best bet. If you would like to help us learn how to make coaching work best in classrooms and staff rooms in every context, then join us on that journey.

A great example of the impact of the Great Teaching Toolkit is the case of Birchgrove Comprehensive school. Find out more about how the Great Teaching Toolkit can help your school too, here.

References

Bambrick-Santoyo, P. (2018). Leverage Leadership 2.0. Wiley.

Blazar, D., & Kraft, M. A. (2015). Exploring mechanisms of effective teacher coaching. Educational Evaluation and Policy Analysis, 37(4), 542–566. https://doi.org/10.3102/0162373715579487

Blazar, D., McNamara, D., & Blue, G. (2022). Instructional coaching personnel and program scalability (499; 21). https://doi.org/10.26300/2des-s681

Deans for Impact. (2016). Practice with purpose: The emerging science of teacher expertise. https://deansforimpact.org/resources/practice-with-purpose/

Goldhaber, D., Krieg, J., & Theobald, R. (2020). Effective like me? Does having a more productive mentor improve the productivity of mentees? Labour Economics, 63, 101792. https://doi.org/10.1016/j.labeco.2019.101792

Knight, D. S. (2012). Assessing the cost of instructional coaching. Journal of Education Finance, 38(1), 52–80. http://www.jstor.org/stable/23259121

Knight, J. (2019). Why teacher autonomy is central to coaching success. Educational Leadership, 77(3). https://www.ascd.org/el/articles/why-teacher-autonomy-is-central-to-coaching-success

Kraft, M. A., 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), 547–588. https://doi.org/10.3102/0034654318759268