top of page
  • Twitter
  • LinkedIn

Instructional coaching: A model for developing agency and supporting retention efforts

Writer: Adam KohlbeckAdam Kohlbeck
Feb 189 min read
ADAM KOHLBECK, CO-FOUNDER, EDUPULSE; DEPUTY HEADTEACHER, BIRKBECK PRIMARY SCHOOL, UK

The United Kingdom has one of the youngest and least experienced teaching workforces in the developed world (OECD, 2020). This represents an experience loss not only for students but also for new teachers coming into the profession, with mentors necessarily taking on their roles earlier in their careers. Part of the theory is that by giving new teachers the best possible start to their career in terms of their development, they are more likely to remain in the profession. Why, then, in the face of such a widespread commitment to mentoring early career teachers, are we still losing so many of them? There is, of course, a myriad of reasons that contribute to this. Martin et al. (2023) produced a report examining many of them. In this article, I will argue that one contributory factor is a dominant behaviourist approach to instructional coaching (as part of the role of a mentor). Although initially empowering, this approach is damaging to the development of agency, which is central to retention.


Agency and instructional coaching 

Before we go any further, it is important to set out my own theoretical standpoint on the efficacy of instructional coaching: I believe that this form of professional development is the most likely way of improving teaching quality. Sims (2017) presents a thorough appraisal of the quality of research evidence available to us in support of the efficacy of instructional coaching. Furthermore, Sims et al.’s (2021) research into the most effective mechanisms of effective professional development enable us to draw clear links between the broad practice of instructional coaching and the most important mechanisms in overall effective teacher development. For example, modelling, rehearsal, social support and the management of cognitive load are all perfectly possible as part of a good instructional coaching programme. 

However, not all instructional coaching programmes are created equal, and some may inadvertently inhibit agency. Priestley et al. (2015) defined agency as teachers actively contributing to the shaping of their work and its conditions. Therefore, it is not something that teachers have but something that they can achieve when they have two things: the necessary knowledge and skills and the conditions to allow for agency. For example, a school has to value creative innovation. Herzberg (1984) concluded that in professional settings, job satisfaction comes from the intangible factors of meaning and purpose, both of which are products of agency. With job satisfaction comes an increased likelihood of remaining in the profession. So, there is clear evidence to suggest that professional agency is key to addressing the retention challenge that we are currently facing as a sector. Before we consider how instructional coaching can be designed to meet this challenge, it is helpful to consider why agency matters to professionals.


Cordingley et al. (2019) make the point that professional identity is not fixed and that it develops and changes over time. Suárez and McGrath (2022) wrote about these changes to professional identity being shaped by the experiences that each individual has, defining it as ‘a collection of beliefs, emotions and perspectives of themselves and their role as teacher, which is in constant development, depends on context and varies between individuals’ (pp. 8–9). This final point – the variation between individuals – is key. It suggests that the approach to the formulation of professional identity is uniquely formed by each individual, meaning that it cannot simply be a case of a mentor transferring their own beliefs and practices onto a less experienced colleague. If this is attempted, it may lead to an initial uptick in effectiveness, but when the less experienced teacher reaches the point of evaluating their practices in the context of their beliefs and existing thought patterns, there will be a distinct lack of material on which to base their evaluation, because they have been basing their practice on the actions of their mentor. This seems to chime with Kraft and Papay (2014), who noted that teacher development slows considerably after an initial period of rapid progress. Suárez and McGrath’s (2022) point about professional identity relating to beliefs and perspectives, and that these change dependent on context, is also key. Emirbayer and Mische (1998), writing about teacher agency, said that teachers need to be able to imagine different ways in which they can respond, with actions triggered by context, and that this involves creatively reconfiguring their knowledge. So, taking the points of Suárez and McGrath (2022) and Emirbayer and Mische (1998) in combination, we can draw a clear link between professional identity and agency, in that both are concerned with the teacher believing and feeling that they can make a positive difference to student learning.


I suggest that one phenomenon that could be contributing to this lack of agency is a heavily behaviourist approach to coaching. Behaviourism is characterised by the belief that only behaviour that is observable and measurable is worth studying (Watson, 1928). When a new teacher joins the profession, they have little prior experience of being a teacher, and therefore less well-developed mental models of how to teach effectively. This creates an environment whereby the observable actions of more experienced teachers are valuable, as they build up an initial concrete understanding of what teaching is. However, a continued focus on the behaviours of successful teachers is less helpful in establishing adaptive expertise (the ability to create new solutions to problems in a complex environment (Frányó and Dorner, 2024)) because, as Mockler (2020) says, professional identity is about ‘combining who I am with what I do’ (p. 23). If the dominant model of coaching that is used is one that focuses heavily on ‘what I do’ without attempting to combine this with ‘who I am’, then we neglect to value intelligent experimentation that underpins agency and cannot hope to develop professional identity. Instead, teachers will perceive what is valued as being compliant with what their mentor does. Crucially, ‘who I am’ refers to the professional ‘I’ and, in this sense, is defined by how I think.


So why is there such an emphasis on a behaviourist approach to instructional coaching? Goodrich (2023) made the point that coaches must develop teachers’ interpretation of what they are seeing (beyond the observable behaviours) and their understanding of why the observable is a positive or negative influence on learning (again, beyond behaviour). Knight (2021) writes at length about coaching needing to address a learning problem. Learning is invisible and so, again, there is a belief that we need to think hard about the unobservable. With such a plethora of seemingly complementary advice about the importance of teacher thinking, it should be easier to achieve a dominant model of instructional coaching that prioritises developing thinking and decision-making over the actions alone. To help to think about why this might be harder than it seems, we return to the opening point of this article – that mentors are now necessarily earlier in their careers.


Klein (2008) wrote about ‘naturalistic decision making’ and explained that in environments characterised by ‘dynamic and uncertain conditions, people were using prior experience to rapidly categorise situations… relying on some kind of synthesis of their experience – call it a schema or a prototype or a category’ (p. 457). This suggests that experts (a term that can be relatively applied to a mentor working with a new teacher) are not conscious of their thought processes in their classrooms. This raises the question of how they can then be in a position to present these thought processes to less experienced teachers. Over time, teachers do develop a greater awareness of why they do what they do, through the repeated exposure to broadly similar cause-and-effect situations. However, mentors are taking on their role earlier in their careers, meaning that they are less likely to be aware of their own decision-making processes and more likely to home in on the observable teacher behaviours that they encourage the less experienced teachers to adopt.


So, to sum up the problem that I believe that our sector is facing, teacher attrition is, in considerable part, due to a lack of professional identity. Professional identity is heavily influenced by agency. Classroom teacher behaviours are easier to observe and therefore easier to coach on, which makes this the focus for lots of instructional coaching programmes. Replicating a more experienced teacher’s behaviours will lead to short-term gains for newer teachers but could inhibit their achievement of agency.


So what can we do about this?

The solution, as I see it, is to build instructional coaching programmes that deliberately develop thinking and subsequent decision-making, as well as a repertoire of techniques. Cottinghatt (2024) wrote about mental models being ‘our knowledge of how things work’. Essentially, mental models help us to simplify what is going on in our environment in a way that triggers an action. This reinforces Klein’s (2008) idea that expert classroom decision-making is unconscious. The problem with which this presents us is that if we want novice teachers to learn from the mental models of their mentors, they need to be able to visualise these models.


One potential solution to this is to map mental models as part of instructional coaching. This process involves the coach and teacher working through the process of the desired goal of a teaching episode, the actions that the teacher took and the concerns that drove that action in the context of the goal and then the effect of the actions on student learning. This facilitates an examination of what the teacher was aware of and driven by prior to making the judgement that preceded their actions. This is important because it enables reflection on whether or not this was the most powerful thing to be concerned with at the time, and how this might necessarily lead the teacher to change their actions. It also allows the teacher space to make a prediction about the likely effect of a different action.


By adopting an approach of surfacing and adapting mental models, we can reveal the decision-making process of the teacher and, by comparing it to the mental model of the same situation that the mentor holds, help them to analyse and choose to change it. Of course, locking in this change still requires practice and there is a strong argument for the use of rehearsal to help to develop the new habit of thinking.


By empowering teachers with the knowledge and understanding of their existing mental model, we can surface the subconscious concerns and, armed with knowledge, we can present them with alternative concerns and help them to make a choice about whether or not a more powerful concern existed. By unpacking thought processes, instructional coaching conversations can get to the route of decision-making and can offer teachers a framework to reconsider their judgements. This approach may lead to the achievement of agency for the teacher, and therefore a stronger sense of professional identity. In turn, this makes it more likely that they will remain in the profession and, in time, be in a position to coach other teachers in the same way, passing the baton of decision-making and going some way to ensuring the future of the profession.


 
 
 

Comments

download (1).png
stepup-4.png

© 2025 EduPulse

bottom of page