Abstract
Supervision in educational leadership has long been framed as an evaluative process. Its purpose, on this familiar view, is to monitor performance, enforce standards, and account upwards. The trouble is that this framing tends to produce compliance rather than growth, and the work of contemporary school leadership demands far more than compliance. Principals today operate in conditions of genuine complexity, with ambiguous problems, competing stakeholder claims, and limited time for the kind of reflective work that complex decisions actually require. This conceptual article presents non-directive coaching supervision in an extended one-on-one setting as an alternative practice of educational leadership supervision. Non-directive coaching supervision is the art and practice of supervision through inquiry, rather than instruction, drawing on theories of coaching, adult learning, reflective practice, and transformative learning. This paper proposes that non-directive coaching supervision is a developmental and dialogic practice in the development of educational leaders’ thinking, professional action, and adaptive expertise. Drawing on the writings of international researchers in educational supervision and leadership coaching, this article proposes the practice of coaching supervision, its effectiveness in the development of educational leaders, and how it is differentiated from related developmental practices such as mentoring and instructional coaching. A conceptual framework for the practice of coaching supervision is proposed with four major components: the foundational conditions for coaching supervision; the supervisory processes; the educational leader's thinking, action, and learning or leadership capacities fostered through coaching supervision; and the outcomes of the educational leader's leadership. The article also discusses the potential compatibility of coaching supervision with accountability, cultural and equity issues relevant to supervision, and research priorities. The central contention is straightforward. If supervision is to contribute meaningfully to leadership development—and not merely to leadership surveillance—then its dominant evaluative framing must yield ground to practices that treat school leaders as the central thinkers in their own professional growth.
1 Introduction: rethinking supervision in educational leadership
Supervision is one of the most valued and least satisfying processes within the educational leadership system; school leaders are rarely satisfied with how it is operationalized. Supervision is traditionally described as a process that serves an organization's controlling function by monitoring work performance, enforcing standards, and reporting outcomes to higher authorities (, p. 4; , p. 12). This controlling function has persisted regardless of reform: developmental aims announce themselves at the start of a reform cycle, and evaluative routines reassert themselves by the end (). Supervision aimed at supporting the school leader is often not how the school leader experiences the process ().
Why has this been so difficult to change? Part of the explanation rests with the scope of the supervision. Over the past two decades, the expectations for school leaders have expanded. Principals are expected to be instructional leaders, change agents, community builders, equity advocates, and system thinkers—sometimes within the same conversation (; ). They are grappling with accountability pressures combined with staffing, fiscal, and technological pressures in a post-pandemic world. Much of this work comprises what , p. 49) memorably describes as “indeterminate zones of practice"—a principal is faced with situations where the rules end, and a judgment made under conditions of uncertainty is the only recourse. Supervision framed around making judgments about a worker's performance against a set of predetermined standards is ill-suited to work where the standards are contested.
Recent literature reflects this incongruence and a call for a paradigm shift for supervision to be a catalyst for leadership growth rather than a mechanism for control (, p. 9; , p. 31). Much has changed in discourse, less so in practice: supervision sessions in many contexts remain appraisal conversations masquerading in developmental jargon. It is the disparity between the talk about and the practice of supervision that partly drives this article. The rhetoric about “a supervision for growth” will continue to sound hollow as long as the realities of that supervision structure reward conformity and penalize honesty, which is necessary for reflection ().
In response to the questions posed by the supervisory reform literature—what models support leadership development, and how does supervision move beyond compliance?—I suggest that one of the most pertinent answers, a model of non-directive coaching supervision, is more often cited than defined in the research literature. Three central questions guide this discussion: first, what are the theoretical and empirical arguments for coaching supervision as a developmental experience distinct from other types of supervision? Second, how does it differ, in stance and effect, from the related practices of mentoring, instructional coaching, and developmental supervision in the Glickman tradition? Third, what are the practical implications of including it within a system of effective developmental practices for school leaders?
By asking these questions, I make the case that non-directive coaching supervision is not merely a gentler form of evaluative supervision; it is supervision as an entirely different practice, with a different theory of how leaders learn, a different understanding of where supervisory power and authority come from, and a different relationship to accountability. The change called for is anything but cosmetic; it challenges our understanding of the purpose of supervisors, the nature of their expertise, and how development and evaluation should be integrated. The contribution lies not in new elements—non-directive coaching, developmental supervision, adult learning, and reflective practice each have established literatures—but in a new application: to my knowledge, no one has applied the non-directive end of the coaching spectrum specifically to the supervision of school leaders—a relationship instead shaped by developmental supervision in the Glickman tradition (Section 2.2) and inspection-based accountability regimes (Section 2.1), neither designed with the seniority, isolation, and decision-making complexity of the school leader in mind. make a related point in their systematic review: existing literature tends to treat coaching as an add-on to supervision rather than a redefinition of it. This is consistent with the absence of any study meeting that review's criteria—published between July 2014 and July 2024—that applies a predominantly non-directive coaching approach to school leader supervision, suggesting the gap identified here is not an artifact of the search itself.
Those coming from the Glickman tradition or cognitive coaching may see this as continuous with their own work rather than a departure from it; I would describe it instead as that same synthesis applied with greater precision to this particular population. Specifically, this article contributes a fully conceptualized version of non-directive coaching supervision that articulates three elements that remain incomplete or implicit in existing models: non-directiveness as a general stance rather than an occasional approach, bracketing from summative evaluation, and respect for the leader's own sense of context. In addition, this article contributes a conceptual model relating the conditions of supervision to leader outcomes, including the processes and capacities in between, and explores the systems question of how coaching and accountability can co-exist without one subverting the other.
The paper is structured as follows: Section 2 presents the history of supervision and the tensions within. Section 3 presents the background of coaching supervision. Section 4 defines non-directive coaching supervision and explores the role of the supervisor as a thinking partner. Section 5 explores the implications for leadership practice. Section 6 posits the conceptual framework. Section 7 explores implications for systemic and policy-level leadership, including cultural and equity implications. Section 8 provides the conclusion with recommendations for research.
2 Supervision in educational leadership: from control to capacity building
2.1 Traditional supervisory paradigms
Educational supervision has bureaucratic roots, and they show. Inspection-type supervisory practices accompanying the rise of mass schooling in the nineteenth and early twentieth centuries reflected the era's emphasis on control, standardization, and efficiency, much as in other workplaces (, p. 47): supervisors oversaw, teachers obeyed, and reports passed upward through the hierarchy. This structure, tied to the need for predictability across a dispersed system, has proven remarkably immune to change—supervision practices first implemented when schools were founded tend to reappear under whatever reform movement is current ().
Today, this reassertion appears in the principal's performance appraisal process, leadership standards, and external inspection, where measurable outcomes and summative judgments predominate (). The international move toward standards-based accountability has, if anything, reinforced external inspection: find that high-stakes testing pushes school leaders toward safe, defensible practices and away from the risk-taking real improvement requires.
Yet the evaluative framing is also doing real work: school systems need ways of identifying serious underperformance, and communities need assurance that schools are accountable. The case against evaluative supervision is not that accountability is unimportant; it is that supervision-as-accountability crowds out supervision-as-development when one encounter must serve both purposes. Despite three decades of rhetorical commitment to “developmental supervision,” most supervisory practice remains predominantly evaluative—the vocabulary has changed faster than the practice.
2.2 Developmental and instructional supervision
Scholars responding to the limits of inspection have offered more developmental alternatives. The Glickman tradition—developmental supervision in its various iterations—represents perhaps the most influential of these (; , p. 112). Its premise is that supervisees operate at different levels of professional maturity, abstract reasoning, and commitment, and that effective supervision adjusts its stance accordingly: directive with a novice teacher struggling with classroom management, collaborative with a mid-career colleague working on curricular adaptation, non-directive with an experienced teacher engaged in self-directed inquiry. The framework is contingent: stance follows diagnosis.
This is a genuine theoretical advance over inspection, shaping supervisory practice for nearly four decades. But two features, as typically enacted, constrain its impact on school leaders. First, the supervisor retains diagnostic authority: it is the supervisor who decides what developmental level the supervisee is operating at, with the supervisee's own sense of their needs treated as one input among others rather than the determining factor. Second, the non-directive end of the continuum, while available in theory, is reached rarely in practice—directive and collaborative stances are more familiar and more compatible with supervisors’ typical time pressures. Non-directive supervision is the developmental ideal everyone endorses but few practice.
Applied to school leaders—rather than teachers, the population for which it was designed—these limitations grow more pronounced. Leaders themselves supervise teachers, lead organizational learning at scale, and make decisions that ripple across an institution; treating them as supervisees in the standard model risks treating them as more junior than they are, and misses the more interesting question: how do experienced professionals keep growing as demands on them increase?
2.3 The formative–summative tension in supervision
A persistent tension runs through the supervisory literature: the difficulty of doing formative and summative work in the same conversation, with the same person, at the same time. Supervisors are routinely expected to support a leader's development and evaluate the leader's performance, often within the same year, sometimes within the same meeting. For decades, research has shown that this pairing tends to inhibit the frankness upon which developmental conversation relies (, p. 56). A leader who has just made a difficult, controversial decision about staff employment, programming, or handling a complaint from a community member will be cautious in conversations with a supervisor who will later judge them on those decisions. The conversations they need most are the ones they are most inclined to avoid.
This is particularly detrimental to the practice of supervising school leaders, because leadership itself requires behaviors that evaluative supervision suppresses: experimenting, admitting uncertainty, testing hunches, articulating underlying assumptions—none easy under evaluative gaze. Supervisory models that overemphasize summative evaluation may, in the worst case, train leaders to perform a leadership style for their supervisors that is quite different from the leadership they actually practice. The supervision becomes a kind of theatre, with one set of decisions made for the supervisor and another, perhaps more honest set, made in the real life of the school.
We cannot simply wish away this tension. Schools and systems need formative and summative supervision; the question is how to arrange this so the two do not work at cross-purposes. Drawing on the literature in coaching in education, one way is to separate formative and summative supervision—either through two different people, or by clearly delineating functions and pre-establishing the conditions of confidentiality (, p. 30; , p. 56; , p. 42). The alternative—leaving the tension implicit and hoping for the best—consistently produces the worst of both worlds: supervision too evaluative to support genuine development, and too developmental in its rhetoric to be experienced as fair evaluation.
2.4 The case for reconceptualization
This journey through the history of supervision models suggests repeated attempts to increase the role of development in supervision—yet each model returns to evaluation and control by the supervisor. This is not the failure of any individual reformer; rather, incremental reforms have been unable to overcome a larger institutional dynamic. Supervision, as currently configured in most systems, is structured to serve accountability, and layering developmental purposes on top of that structure tends, predictably, to produce hybrid arrangements in which the accountability purpose dominates in practice.
A more substantive reconceptualization may therefore be necessary. The argument advanced here is that the resources for such a reconceptualization lie partly outside the traditional supervision literature—in the more recently developed body of work on coaching, adult learning, reflective practice, and transformative learning. These traditions share a commitment that the supervision literature has, on the whole, struggled to enact: treating the professional being supervised as the central thinker in their own development, with the supervisor in a structuring rather than prescriptive role. The next section develops these theoretical foundations.
3 Coaching supervision as a theoretical and developmental practice
3.1 Coaching in educational leadership
Coaching has become one of the most actively researched professional development practices in education over the past two decades (). Definitions vary, but most converge on a few common elements: coaching is a structured, sustained, one-to-one relationship oriented toward the development of the person being coached, conducted through dialogue rather than instruction, and focused on the coachee's own goals, dilemmas, and growth (, p. 32; , p. 18; , p. 6). It is distinguished from training by its individualized, dialogic character; from mentoring by not relying on the more experienced party's expertise or personal example as a primary resource; and from therapy by its present-and-future, professional rather than psychological, orientation.
Coaching has been used in education in multiple forms, including instructional coaching, which targets classroom practice (), and leadership coaching, directed toward serving and aspiring principals (; , p. 62). Across these forms, the empirical evidence is encouraging: coaching has been linked to gains in instructional focus, communication, strategic decision-making, and professional confidence among school leaders (, p. 12; , p. 235; , p. 75; ).
Two features distinguish coaching from the dominant supervisory model: individualization, since coaching engages the specific dilemmas of an individual leader rather than common content across a group; and source of expertise, since coaching relies on the coachee's own knowledge of their situation rather than the trainer's content expertise. The coach is an expert in process, not in prescribing solutions—reflecting the view that experienced professionals already hold much of what they need, requiring only time and structure to surface and use it.
3.2 Adult learning and reflective practice
Adult learning theory is a key contributor to the rationale for coaching supervision. Knowles and colleagues’ concept of andragogy holds that adult learners direct themselves within the learning process, with past experience integral to any learning situation (, pp. 60–65); they learn best when they see immediate relevance to their work, when learning centers on problems rather than content, and when their autonomy and experience are respected rather than treated as empty vessels to fill with knowledge.
Reflective practice theory provides a complementary foundation. argued that professional practice in complex domains cannot be fully captured by technical-rational models of problem-solving (pp. 21–26): practitioners operate in “indeterminate zones"—situations marked by uncertainty, complexity, value conflict, and uniqueness (p. 42)—where they frame problems, test interpretations, and reflect on the consequences of action rather than apply pre-existing knowledge to well-defined problems. Schön called this “reflection-in-action” and “reflection-on-action,” the core mode through which professional expertise develops in complex domains (pp. 54, 68).
School leadership is paradigmatically such a domain. Few of the most important decisions a principal makes have textbook answers—should a struggling teacher receive more support or be eased toward the door, or is a new initiative working or just generating activity? Each requires framing the problem, weighing competing considerations, acting under uncertainty, and revising one's thinking in light of what the action reveals. Coaching conversations create deliberate space for this kind of reflective work: open-ended questions, paraphrasing, and tolerance for silence slow the intuitive, rapid thinking that drives many leadership decisions, making leaders’ thought processes visible to them, often for the first time. Over time, the habits of reflection a coaching conversation provides are theorized to become internalized (, p. 4)—leaders gradually beginning to pose for themselves the questions a coach would ask.
3.3 Transformative learning and critical reflection
Another theoretical thread that needs to be woven into the explanation is that of transformative learning, as discussed by Mezirow and his colleagues (). While theories of reflective practice address how professionals make sense of indeterminate situations, the theory of transformative learning addresses how adults change the assumptions, the frame of reference, the lens through which they view experiences. Transformative learning does not simply mean that a person has acquired new knowledge or skills; it means that, often uncomfortably, a person has come to recognize the limitations of the frame through which he or she views a situation and has adopted a more discerning lens.
This is a significant distinction for school leaders. Many of the developmental moves discussed above involve revisioning rather than adding new techniques—a conflict-avoidant principal realizing their staff needs more candour, or a leader realizing that discipline practices grounded in a deficit mindset are inconsistent with their own values. These are revisions in perception rather than technique, and they often produce a strong, positive ripple effect on practice. It is precisely these revisioning moments that skillful coaching conversations are likely to support, through the safe space the coach creates for inquiry and provocative questioning.
3.4 Non-directive coaching: a distinct supervisory stance
Coaching approaches span a spectrum: directive, where the coach offers advice and solutions; guided or facilitative, a middle ground combining questions with suggestions; and non-directive, also referred to as pure coaching, which relies almost exclusively on questions, listening, and reflection. Here, the coach offers no advice but instead supports the coachee in constructing their own analysis and conclusions (, p. 16). The lines can be blurred, and effective coaches will probably move along the continuum within each coaching session, according to the needs of the coachee and the issue at hand.
This raises a boundary question worth addressing directly. If non-directive coaching offers no advice, what should be made of practices described later in this article (Section 4.2)—offering an observation, or naming an alternative the leader has not considered? The distinction is one of function, not form: a non-directive observation is offered as material for the leader's own analysis, something to take up, set aside, or argue against, rather than something the leader is expected to act on because the supervisor believes it is correct. The relevant question is whether the leader's response is invited to be evaluative—does this fit my situation, and why—rather than compliant: the supervisor said to do this, so I will. Guided or facilitative coaching, by contrast, pairs suggestions with an expectation that the coachee will act on them. A supervisor practicing non-directive coaching supervision may, on occasion, name something the leader has not raised, but does so rarely and tentatively, with authority over the idea left explicitly with the leader.
What makes the non-directive end of the continuum particularly relevant to the supervision of school leaders is the combination of three features. School leaders typically already possess substantial professional expertise and contextual knowledge; that knowledge is not easily accessible to an external supervisor, who lacks the situational detail and the relational history; and sustainable leadership development depends on building the leader's capacity for independent judgment, not on the leader's deference to external prescription. Non-directive coaching honours each of these realities, treating the leader's expertise as the primary resource, respecting that contextual judgment is irreducibly the leader's, and developing the leader's reasoning capacity rather than substituting for it.
The non-directive approach is not passive; it requires considerable expertise. The supervisor must resist the natural tendency to advise, especially when most confident about what the leader should do, opening exploration rather than guiding the leader toward a prescribed resolution, and tolerating silence rather than rushing to fill it. Non-directive coaching is deceptively difficult and is often hard for people primarily trained in evaluative supervision to adopt without considerable development.
3.5 Distinguishing coaching supervision from related practices
Coaching supervision is related to several practices in educational leadership development, and it is useful to distinguish between them. Mentoring is the closest relation, also one-to-one, ongoing, and developmental. But the difference lies with the role of the more experienced party. Mentors draw on their own lived experience and personal example as a primary resource for providing guidance; discussions between mentor and mentee often focus on what the mentor did, would do, or has learned. In contrast, coaching supervision purposefully de-emphasizes the supervisor's own experience and highlights the leader's thinking. The shift is not absolute—an experienced coach occasionally shares relevant experience, but the default direction of attention differs (, p. 28).
Instructional coaching, as described in much of the North American literature, is more directive and more directly engaged with classroom practice than coaching supervision (), typically involving a cycle of modeling, observation, and feedback tied to a specific practice. Applied to leaders rather than teachers, it tends to look like structured expert advice on instructional leadership—valuable, particularly for novice leaders, but not what this article means by coaching supervision.
Developmental supervision in the Glickman tradition presents a non-directive stance as one of the options on a continuum but retains the authority of the supervisor to determine the agenda, diagnose the developmental level, and, ultimately, evaluate. Coaching supervision is different in that the supervisor takes a non-directive stance predominantly rather than as one option within a range; the relationship is intentionally removed from summative evaluation rather than merging the roles in the supervision; and the leader retains the ownership of the agenda, analysis, and action steps. Table 1 distinguishes non-directive coaching supervision from mentoring and instructional/directive coaching, while Table 2 details the proposed framework's own four components.
Table 1
| Dimension | Mentoring | Instructional/directive coaching | Non-directive coaching supervision |
|---|---|---|---|
| Primary stance | Experienced practitioner shares expertise and example | Coach provides content advice and modelling | Supervisor uses inquiry and reflection to support leader-generated insight |
| Source of expertise | Mentor's prior experience | Coach's content knowledge | Leader's contextual knowledge; supervisor's process skill |
| Ownership of agenda | Often shared, mentor-influenced | Often coach-led | Leader retains ownership |
| Relationship to evaluation | Typically separated | Often separated; sometimes blended | Deliberately bracketed from summative evaluation |
| Primary developmental aim | Transmission of expertise; induction | Acquisition of specific practices | Reflective judgment, professional agency, adaptive expertise |
| Typical duration | Often time-limited induction period | Cycle-based, problem-of-practice focused | Sustained engagement across leadership trajectory |
| Principal limitations | Inherits the mentor's own assumptions and blind spots; risks passing on outdated practice in a rapidly evolving environment; particularly unfit for leaders whose context differs substantially from the mentor's | Fosters dependence on the coach's advice rather than the leader's own judgment; less effective for genuinely novel or unclear problems without an established “right” practice to draw on | Requires considerable skill and restraint from the supervisor; slower to show results than more directive approaches; requires sustained commitment and resourcing that some systems may not be able to mobilize (Section 7.1) |
Distinguishing non-directive coaching supervision from related developmental practices in educational leadership.
Table 2
| Framework component | Key elements |
|---|---|
| Foundational conditions | Trust, psychological safety, clarity of purpose, separation from summative evaluation, procedural confidentiality, sustained relational engagement over time. |
| Supervisory processes | Reflective dialogue, powerful open-ended questioning, active and disciplined listening, joint inquiry into leadership dilemmas, suspension of supervisor judgment. |
| Leadership capacities | Critical thinking, self-efficacy, adaptive expertise, reflective judgment, emotional regulation, metacognitive awareness, clarity of professional values. |
| Leadership outcomes | Improved leadership practice, strengthened teacher development and school culture, sustainable leadership identity, contributions to school improvement, leader wellbeing and retention. |
Core elements of the proposed conceptual framework for non-directive coaching supervision.
4 Non-directive coaching supervision: a reconceptualization of supervisory practice
4.1 Defining non-directive coaching supervision
Non-directive coaching supervision is a continuing, one-on-one relationship in which the supervisor adopts the stance of coaching—inquiry, reflection, and sense-making—rather than evaluation or direction. The supervisor's expertise is conversational: the ability to ask questions, listen, and create conditions for reflection, rather than expertise in the content of the leader's work. The leader remains the content expert, knowledgeable about their school, staff, and current challenges; the supervisor is the process expert who enables examination of that knowledge.
Respect, genuine curiosity, and patience underpin this work. The supervisor builds a trusting relationship over time, asks questions that open thinking rather than close it, and stays engaged long enough to see patterns of practice rather than isolated incidents. Clarity about confidentiality is essential throughout—what will and will not be reported to evaluation authorities, and under what circumstances confidentiality might be broken.
These features complement each other: trust allows honest reflection, and rigorous questioning requires respectful listening. The absence of any one diminishes the others, as the next section discusses in more detail.
4.2 The supervisor as thinking partner
The role of the supervisor in coaching supervision is that of a thinking partner, not an evaluator—a shift of role, not merely of terminology. The thinking partner role implies a skillful engagement: asking powerful questions, making explicit what is implicit, and—occasionally and tentatively—naming an observation the leader has not raised, offered as material for the leader's own analysis rather than a recommendation to be acted upon (Section 3.4). It is an engaged stance, requiring stamina, generosity, and a willingness to be useful without providing the answer.
What makes this difficult, and valuable, is restraint. Experienced supervisors have earned ideas about what a leader should do, but sharing them too readily short-circuits the leader's thinking and creates a dependency in which the leader's own thinking atrophies. By abstaining from solutions, the supervisor creates space for the kind of independent problem-solving theorized to support the development of metacognition, judgment, and confidence over time (, p. 28).
What is happening is that authority is shifted rather than removed: it derives not from the supervisor's position to judge or command, but from the skill of structuring a conversation that elicits insights the leader could not have generated alone. This authority is more challenging to command, since it is earned in every conversation rather than granted by hierarchy—and it is this less hierarchical quality that lets leaders interact more openly with thinking partners than with evaluators, enabling developmental work to occur.
4.3 Enhancing critical thinking and reflection
Non-directive coaching supervision tends to target the leader's thinking. By posing reflective questions, the supervisor asks the leader to consider their assumptions, seek evidence, and examine the logic of their conclusions. Effective questions are open rather than leading, inviting the leader to think rather than supplying the right answer, and considering the broader patterns of a leader's practice rather than the issue in isolation. A supervisor might ask: “What are you assuming about the staff's reaction?” or “What would you need to see to be confident the new approach is working?” Such questions invite the leader to seek evidence and challenge habitual responses—and over time become internalized, with the leader beginning to ask these questions of themselves.
These experiences strengthen a leader's decision-making and reflective judgment. Recent leadership literature points to the importance of reflective, critical thinking for effective leadership in complex educational systems (, p. 11): practicing reflective inquiry strengthens the skill, much as a supervisory conversation functions as a kind of mental exercise the leader never fully stops needing. The practice of supervision is about strengthening the leader's own judgment, not creating external compliance.
4.4 Trust, confidentiality, and the conditions of reflection
All of the above rests on easily assumed and difficult to earn premises. Trust is primary among them. Leaders must trust that what they share will not be used against them, and that the supervisor seeks their growth rather than their compliance. They must also trust that their time with the supervisor is worth the time given, believing in the skill and integrity of the supervisor. These cannot be assumed. They must be developed through consistent behaviors of the supervisor, being reliable, demonstrating competence, being willing to admit that one does not know it all, and protecting confidentiality.
Confidentiality issues must be addressed with clear procedures. In many systems, the supervisors of school leaders have access to evaluation data and report to those making decisions about leaders’ futures; if a supervisor also serves as a coach, clear agreements about data sharing are essential, and the coaching role cannot factor into evaluation or decisions about the leader's future. Without such procedures, reflective conversations risk being mistaken for evaluations.
This may seem like a procedural detail. But in reality, it is what makes the model viable or not. A coaching supervision relationship without protected space is like an interview without confidentiality: the interview takes place, but it generates a certain kind of response.
5 Implications of coaching supervision for leadership practice
5.1 From reflection to practice
Simply asserting that coaching supervision produces more reflective leaders is not sufficient to establish its value, nor is the further claim that this benefits teaching only indirectly. Some might argue that if the intended outcome is student learning, coaching should target teachers directly rather than leaders. This skepticism does not hold up. While a growing body of research shows reflective supervision brings with it changes in a leader's practice in the form of instructional focus, communication, and strategy (, p. 235; , p. 75), coaching can also identify unexamined theories of action and expose them to scrutiny; surface early warning signs of unproductive patterns such as avoidance or over-reliance on familiar strategies; and help leaders rehearse new approaches to difficult conversations in advance, so reflective insight more easily translates into practice. Leaders, moreover, create the conditions that make teacher coaching possible in the first place—protecting time and resources for it, and shaping whether a school is a place where a teacher can safely be vulnerable about their practice. Coaching leaders is therefore an investment in the conditions that enable coaching of teachers, not a substitute for it; high-quality teacher coaching, when well-resourced, remains invaluable in its own right, and the choice between the two is not zero-sum.
Coaching supervision meets a principal's need for guidance in ambiguity. Difficult issues—a declining staff member's performance, a school turnaround that risks alienating the staff who built it—rarely come with a playbook. These demand a leader's capacity to hold competing dynamics in tension, act under uncertainty, and think flexibly; coaching conversations help develop both that flexibility and emotional regulation.
5.2 Indirect impact on teaching and learning
While coaching supervision occurs with the leader as the direct recipient, the practice may, in theory, have ripple effects. According to the leadership literature, effective school leaders influence student outcomes indirectly through teachers, school culture, and the instructional milieu (, p. 13). If supervision's reflective work helps a leader develop teachers and build a trusting professional culture, better outcomes for teaching and learning are plausible, though not guaranteed. Consistent with school improvement research, find that exemplary school leaders model lifelong learning, inspiring a stronger culture of teacher development and collaboration in their schools (p. 134).
5.3 Professional agency and leadership identity
Promoting autonomy in learning is assumed to have the effect of supporting the leader's sense of agency in their professional path. A school leader receiving coaching supervision might emerge with a reinforced sense of their own values, a more vibrant enactment of their leadership will, and a greater assurance of their direction in times of turbulence (). This is no exercise in nebulous self-improvement: amid accountability pressures and rapidly changing policy, many school leaders risk feeling like instruments of others’ decisions rather than autonomous professionals.
Coaching supervision offers a counterbalance: the leader's continual interpretation and choice-making keep their identity robust, treating directives as something to be filtered through their own judgment rather than simply obeyed. Many leaders, because of this ongoing coaching, describe themselves as anchored in their own values, picking their battles, and able to tell the difference between what can be changed and what must be worked around. It is here, within the space of supervision, that leadership identity emerges, not in a performance review.
5.4 Wellbeing and the sustainability of leadership
A further implication, not emphasized as much in the literature on supervision but equally important in practice, is the well-being of the leader. The principalship is a highly stressful role that leads to role overload, sleep disturbance, and higher turnover (, p. 19). Some of this is systemic and beyond what coaching alone can resolve; some is more manageable. School leaders bear the weight of many of their decisions alone because few opportunities exist to explore the challenging facets of the work. Coaching supervision offers such an opportunity. A confidential, regular, skilled conversation about the realities of the work (the doubts, the second-guessing, the things that cause sleepless nights) is plausibly one contributor to the well-being a long-term role in educational leadership requires—a connection consistent with longitudinal evidence that job resources, including supportive relationships, are associated with reduced burnout among school principals over time ().
Coaching is not a substitute for therapy, and supervisors are not trained to diagnose or treat more serious mental health issues leaders may face. That said, within its scope, the reflective space coaching supervision provides can contribute to the well-being of professionals in demanding roles, offering an alternative to leaders processing the hardest parts of their work alone.
6 A conceptual framework for non-directive coaching supervision
This section synthesizes the previous sections into a conceptual framework, linking the theoretical basis in Section 3, the practices in Section 4, and the implications in Section 5. It serves both as a conceptual map for researchers and a practical heuristic for supervisors, leaders, and system designers.
The framework consists of four interconnected components:
Foundational Conditions: trust, psychological safety, clarity of purpose, separation of role from summative evaluation, confidentiality of procedures, and sustained relationship over time.
Supervisory Processes: reflective dialogue, powerful questioning, active listening, and deferring judgment from the supervisor in exchange for curiosity about the leader's answer.
Leadership Capacities: critical thinking, self-efficacy, adaptive expertise, reflective judgment, emotional regulation, metacognition, and clarity of professional values.
Leadership Outcomes: improved leadership practice, teacher development and school culture, leadership identity that carries over to the next year, contribution to school improvement, and leadership well-being and retention.
The link between foundational conditions and supervisory processes is fairly direct: without trust and confidentiality, honest disclosure is unlikely, since a leader who fears that a coaching conversation may surface in a performance review has reason to withhold the doubts that reflective dialogue depends on (Section
4.4). Without these conditions, supervisory processes would not disappear, but their content would shift toward safer material.
The link to leadership capacities is more indirect. Sustained dialogue creates repeated opportunities to practice specific cognitive moves—generating alternative interpretations, tolerating ambiguity—that these capacities depend on (Section 3.2), consistent with adult learning theory's claim that such capacities develop through practice rather than instruction alone ().
The link between capacities and outcomes is the most hypothetical: stronger reflective judgment does not guarantee an improved school culture, since variables such as staff stability and resources can interfere. The more defensible claim is that capacities make outcomes more achievable, not that they alone produce them. Finally, the loop closing from outcomes back to foundational conditions works through a social mechanism: visible progress sustains continued investment of trust and time; its absence erodes it. This is why the framework is drawn as a cycle rather than a line. Each of the links just described, however, has so far been stated only descriptively; restating them formally makes clearer what would actually need to be tested.
Stated formally:
Proposition 1: Stronger foundational conditions predict more candid supervisory dialogue.
Proposition 2: Sustained non-directive dialogue predicts gains in leadership capacities, mediated by repeated practice of underlying cognitive moves.
Proposition 3: The capacities–outcomes relationship is moderated by contextual factors such as staff stability and resources.
These are offered as starting points for the empirical work discussed in Section 8.
The framework treats supervision as a developmental system rather than an isolated activity: the supervisory conversation sits within a constellation of conditions and processes that must align for coaching supervision to work, and the framework indicates what counts as evidence that it is. These propositions also generate diagnostic logic: if conditions align and processes are well-enacted but capacities are not developing, something is amiss; if capacities develop but predicted effects on teaching, culture, and learning do not follow, either the theory linking capacities to outcomes is flawed, or the broader context is blocking what those capacities would otherwise produce. The framework, in short, yields testable claims, not merely descriptive ones.
6.1 Toward operationalizing the framework’s constructs
The framework's value as a generator of testable propositions depends on the constructs themselves being more than evocative labels. Critical thinking, self-efficacy, adaptive expertise, reflective judgment, professional agency, leadership identity, and metacognition are each established constructs in the broader psychological and educational literature, but their application to this framework requires some indication of how a researcher or practitioner might recognize them in the practice of coaching supervision, rather than simply asserting that coaching supervision produces them. The formal propositions introduced in Section 6 specify the relationships to be tested; the indicators below address the further question of how the constructs within those relationships might be observed or measured.
One model to study is work on the development and validation of a psychometric scale tailored to assess resilience among school principals, rather than a scale built for organizational psychology in general. Other cases may include reflective judgment, as demonstrated by a principal's increasing tendency to generate multiple interpretations of a situation before settling on a pattern that could be investigated through transcripts of coaching conversations or via a self-report instrument. Adaptive expertise, or the willingness to depart from familiar solutions in novel situations, can be investigated via interview questions or journal entries that track how a leader's thinking departs from established routines over time. Professional agency may also draw from similar constructs already used in coaching research: identified self-efficacy, role clarity, and identity formation as correlates that could be incorporated into measures of agency in future studies.
These illustrations are offered as directions, not as completed measurement models. The framework does not currently specify validated instruments for each construct, and developing or adapting such instruments is precisely the kind of empirical work called for in Section 8. What can be said now is that the constructs are not, in principle, unmeasurable; they are underspecified in their current form, and the path toward specification runs through behavioral indicators, existing adjacent instruments such as resilience inventory, and the kind of longitudinal, mixed-methods designs discussed later in this article. Until that empirical work is carried out, the relationships proposed in Figure 1 should be read as hypotheses the framework generates, rather than as findings it reports.
Figure 1
7 Implications for supervisory systems and leadership development
7.1 Designing supervisory roles and structures
At the system level, the central question is whether coaching and evaluation should be combined in one person or split between two. If combined, bracketing—the disciplined separation of developmental conversation from evaluative judgment—becomes essential, requiring skill from the coach-supervisor, the leader's trust that the bracket will hold, and procedural visibility to verify it. Splitting the roles—an outside coach and a separate evaluating line manager—reduces role conflict but creates a more complex, costlier system. Jurisdictions investing in principal coaching have tended toward this two-person model (, p. 72; , p. 237; , p. 96).
Yet the case for separation is not absolute. Additional relationships can deepen the isolation many leaders report (Section 5.4), and a long-trusted supervisor may sometimes enable more honest conversation than an unfamiliar coach, even one formally separated from evaluation; smaller systems may also lack capacity for two distinct roles. This does not dissolve the earlier case for separation (Section 2.3); it means that case rests on circumstances and risk, not a universal rule. Separation is preferable where evaluations affect job security and a system can sustain it; integration may suit smaller systems or contexts where trust is already established. Either way, developmental conversations must be explicitly protected from evaluative decisions.
This tension may be more than philosophical. found that most coaches in the studies they reviewed functioned in a non-supervisory role, treated as critical to leaders’ sense of safety in disclosing openly—yet found the presence or absence of a supervisory role made no difference to that sense of safety. Perhaps the explicit bracketing agreement, not the formal separation of roles, is the bigger factor; if so, resource-constrained systems have more room than this article's default position assumes to choose their own path, so long as the bracketing is explicit and honored (Section 8).
Whichever design is chosen, supervisor skill cannot be sidestepped. Non-directive coaching does not develop spontaneously from leadership experience; assuming effective former principals make effective coaches is among the most reliable causes of underperforming coaching programmes. Sustained training, supervised practice, and access to coaching supervision for coaches themselves distinguish coaching-by-name from coaching-in-substance.
These features are costly, harder to sustain outside well-resourced systems. found that finding time for coaching, and the cost of training and retaining coaches, were the most frequently cited barriers—a problem sharpened in centralized systems, where separating coaching from evaluation may require an entirely new position rather than reallocating existing staff. Two scalable alternatives are worth naming, though neither has been tested at scale for this model: group or peer coaching among trained school leaders (), and phased implementation beginning with leaders in their first one to three years in post, who stand to gain the most and hold the least power to resist its introduction.
7.2 Policy and accountability
The case for coaching supervision is sometimes misread as a case against accountability. It is not: accountability mechanisms exist for legitimate reasons, and communities and taxpayers have reasonable interests in how publicly funded institutions perform. The argument here is more specific—that developmental and accountability functions should be differentiated, and that genuine reflective practice requires spaces protected from immediate evaluative consequence. Well-designed accountability and well-designed coaching supervision can coexist; the combination is harder to achieve than either alone, but achievable.
Several policy levers could support such integration: identifying coaching supervision as a formalised, resourced part of induction and mid-career development rather than an optional add-on; earmarking funds for coaching with safeguards against their being folded into evaluation work as budgets tighten; including coaching competencies as core requirements in professional standards for supervisors; and having inspection frameworks recognize reflective supervisory practice as a feature of well-functioning schools. None of these levers will do the work alone, but together they can shift the institutional gravity that now drags supervisory practice toward the evaluative pole.
7.3 Cultural and equity considerations
Coaching supervision draws substantially on traditions developed primarily in Anglophone professional contexts, and its non-directive stance reflects assumptions about authority, deference, and disagreement that are not universally shared—meaning thoughtful adaptation, not inapplicability, is what's required elsewhere. In contexts where school leaders may be reluctant to articulate uncertainty or disagreement to a more senior figure, the psychological safety non-directive coaching requires has to be established with particular care, over a longer time horizon. Cultural humility on the part of the supervisor matters more than is sometimes acknowledged; assumptions about what “genuine reflection” looks like, drawn from one professional tradition, can suppress the actual reflective work happening in another.
This consideration is not theoretical. In centralized systems, where school leaders answer to a ministry of education and professional hierarchy carries real weight, non-directive coaching supervision runs against established norms in at least two ways: withholding advice may read as evasion where guidance from a senior figure is the norm, and where disagreeing with a superior carries risk, leaders may perform the reflection they believe a supervisor wants rather than engage in it honestly—echoing the theatre of compliance critiqued earlier (Section 2.3). Neither problem warrants abandoning the model, but both suggest a coaching stance cannot be carried unchanged across cultures: more time may be needed to build trust, the shift from directive to non-directive may need to happen more gradually, or coaches may need training in how local norms around hierarchy and disagreement shape the relationship. What should not be lost, however, is respect for the leader as the central thinker in their own development.
Equity counts. High-quality coaching supervision is not always available in all schools; it is more likely to be offered to highly-resourced or high-profile leaders than to those working in under-resourced or more remote places (, p. 240)—a pattern that risks widening the very inequalities it should help close. If coaching supervision is a developmental practice and not a privilege, its availability needs to reflect that. Rural schools, schools serving historically underserved communities, and schools led by principals from underrepresented groups have at least as strong a claim on coaching supervision as schools that already attract substantial institutional investment.
8 Conclusion and directions for future research
The arguments I have presented in this article suggest that non-directive coaching supervision is a serious alternative to the evaluative approach that has long dominated supervisory practice in education. I am not suggesting that evaluative supervision should be eliminated; accountability remains valid, and systems still need ways of identifying serious underperformance. I am suggesting that developmental supervision needs its own theoretical and structural space, distinct from the accountability functions to which it has too often been subordinated. Coaching supervision, grounded in coaching theory, adult learning, reflective practice, and transformative learning, provides one well-developed model for that space.
The four-component framework advanced in Section 6 is intended both as a synthesis of the argument and as a tool for further work. It identifies foundational conditions, supervisory processes, leadership capacities, and leadership outcomes as the constituent elements of coaching supervision and treats their relationships as recursive rather than linear. The framework is offered with appropriate tentativeness. It is a conceptual proposal, not a validated model, and the empirical work needed to test and refine it has yet to be done at scale.
The analysis points to several directions for further research. These priorities call for specific, longitudinal designs, since the effects of coaching supervision likely develop over years rather than months. A longitudinal mixed-methods design with self-report measures of the proposed leadership capacities, alongside periodic interviews with the leaders, may provide insight into how these capacities develop over the course of a coaching relationship. A direct comparison of school leaders who experience coaching that is organizationally separate from evaluation with those who do not would help resolve the tension between and over whether role separation or procedural bracketing more effectively protects leaders’ sense of safety. Such research would involve developing measures for constructs such as reflective judgment and adaptive expertise and then testing whether the model's proposed pathways hold empirically. Cross-cultural comparative research using a common protocol across centralized and decentralized systems would highlight which aspects of the model transfer between contexts. Comparative effectiveness research comparing one-on-one coaching to group or peer coaching could reveal more efficient options for resource-constrained systems.
Other pressing research questions include how a supervisor's own capacity to coach is actually developed; which training designs are effective; and how a supervisor's practice can be assessed without falling back into the evaluative logic the practice is trying to escape. Research is also needed on how to successfully integrate coaching with accountability, whether role bracketing works or fails, how to prevent the coaching relationship from being conflated with the evaluation relationship, and how to make bracketing feel genuine to the leader rather than nominal. Research is also needed on systems that have made different design choices, and on how the non-directive stance is conceived and applied in educational traditions beyond the Anglophone contexts in which most existing literature originates.
Moving supervision beyond compliance is not a small undertaking: it requires shifts in how supervisory roles are designed, how supervisors are prepared, and how supervision relates to accountability. These shifts will not happen quickly, but the potential payoff—a leadership workforce that is more reflective, more agentive, and better equipped for the genuine complexity contemporary schooling presents—is substantial. Reimagining supervision through non-directive coaching is one route, not the only one, but one with merit, whose theoretical foundations and emerging evidence base warrant the attention this article has tried to give it.
Ethic statement
Not applicable, as this is a conceptual article involving no human participant research.
Statements
Author contributions
SA: Conceptualization, Writing – original draft, Writing – review & editing.
Funding
The author(s) declared that financial support was not received for this work and/or its publication.
Acknowledgments
The author thanks colleagues at the College of Education, Abu Dhabi University, for their constructive feedback on earlier drafts. The author also acknowledges the use of an AI language model (Anthropic's Claude) to assist with literature synthesis and to provide language-proofing suggestions in a report; all citations were verified by the author, and the author is solely responsible for the final argument and content of the manuscript.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was used in the creation of this manuscript. The use of an AI language model (Anthropic's Claude) to assist with literature synthesis and language proofing suggestions in a report.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
AguilarE. (2013). The Art of Coaching: Effective Strategies for School Transformation. San Francisco, CA: Jossey-Bass.
2
BushT. (2013). Instructional leadership and leadership for learning: global and South African perspectives. Educ. Change17 (Suppl. 1), S5–S20. 10.1080/16823206.2014.865986
3
ChenJ. (2023). Conceptualising principal resilience: development and validation of the principal resilience inventory. Educ. Manag. Adm. Leadersh.53, 561–58110.1177/17411432231186294
4
CollinsC.MurphyR.BrownM. (2025). The power of coaching in the professional learning and development of school leaders: an ecological framework and critical insights from a systematic review. Front. Educ.10, 1601455. 10.3389/feduc.2025.1601455
5
CostaA. L.GarmstonR. J. (2016). Cognitive Coaching: Developing Self-Directed Leaders and Learners. 3rd Edn. Lanham, MD: Rowman & Littlefield.
6
FullanM. (2014). The Principal: Three Keys to Maximizing Impact. San Francisco, CA: Jossey-Bass.
7
GlickmanC. D. (1985). Supervision of Instruction: A Developmental Approach. Boston, MA: Allyn and Bacon.
8
GlickmanC. D.GordonS. P.Ross-GordonJ. M.SolisR. (2023). SuperVision and Instructional Leadership: A Developmental Approach. 11th Edn. Boston, MA: Pearson.
9
HargreavesA.FullanM. (2012). Professional Capital: Transforming Teaching in Every School. New York, NY: Teachers College Press.
10
KimS.GhoshR.PoellR.MaltbiaT. E. (2025). Editorial: advancing coaching scholarship. Front. Psychol.16, 1544495. 10.3389/fpsyg.2025.1544495
11
Kirişçi-SarıkayaA.Fayda-KınıkF. S. (2025). Leadership identity in schools: a qualitative inquiry with school administrators. Ulusal Eğitim Akademisi Dergisi (UEAD)9, 185–209. 10.32960/uead.1749120
12
KnightJ. (2007). Instructional Coaching: A Partnership Approach to Improving Instruction. Thousand Oaks, CA: Corwin.
13
KnowlesM. S.HoltonE. F.SwansonR. A. (2015). The Adult Learner: The Definitive Classic in Adult Education and Human Resource Development. 8th Edn. London: Routledge.
14
LeithwoodK.HarrisA.HopkinsD. (2020). Seven strong claims about successful school leadership revisited. Sch. Leadersh. Manag.40, 5–22. 10.1080/13632434.2019.1596077
15
LochmillerC. R. (2014). Leadership coaching in an induction program for novice principals: a 3-year study. J. Res. Leadersh. Educ.9, 59–84. 10.1177/1942775113502020
16
MarshH. W.DickeT.RileyP.ParkerP. D.GuoJ.BasarkodG.et al (2023). School principals’ mental health and well-being under threat: a longitudinal analysis of workplace demands, resources, burnout, and well-being. Appl. Psychol. Health Well-Being15, 999–1027. 10.1111/aphw.12423
17
MezirowJ. (2000). Learning as Transformation: Critical Perspectives on a Theory in Progress. San Francisco, CA: Jossey-Bass.
18
NicolauA.CandelO. S.ConstantinT.KleingeldA. (2023). The effects of executive coaching on behaviors, attitudes, and personal characteristics: a meta-analysis of randomized control trial studies. Front. Psychol.14, 1089797. 10.3389/fpsyg.2023.1089797
19
OECD (2013). Leadership for 21st Century Learning. Paris: OECD Publishing. 10.1787/9789264205406-en
20
PassmoreJ. (2016). Excellence in Coaching: The Industry Guide. 3rd Edn. London: Kogan Page.
21
RobertsonJ. (2008). Coaching Educational Leadership: Building Leadership Capacity Through Partnership. London: Sage.
22
SchönD. A. (1983). The Reflective Practitioner: How Professionals Think in Action. New York, NY: Basic Books.
23
SergiovanniT. J.StarrattR. J. (2007). Supervision: A Redefinition. 8th Edn. New York, NY: McGraw-Hill.
24
TyackD.CubanL. (1995). Tinkering Toward Utopia: A Century of Public School Reform. Cambridge, MA: Harvard University Press.
25
van NieuwerburghC. (2012). Coaching in Education: Getting Better Results for Students, Educators, and Parents. London: Karnac Books.
26
WiseD.CavazosB. (2017). Leadership coaching for principals: a national study. Mentor. Tutor. Partnersh. Learn.25, 223–245. 10.1080/13611267.2017.1327690
27
ZepedaS. J. (2017). Instructional Supervision: Applying Tools and Concepts. 4th Edn. New York, NY: Routledge.
Summary
Keywords
capacity building, coaching supervision, educational supervision, leadership development, non-directive coaching, professional agency, reflective practice, school leadership
Citation
Alreiahy S (2026) From evaluation to empowerment: non-directive coaching supervision as a catalyst for school leadership development. Front. Educ. 11:1890429. doi: 10.3389/feduc.2026.1890429
Received
25 May 2026
Revised
20 July 2026
Accepted
24 July 2026
Published
12 August 2026
Volume
11 - 2026
Edited by
Sana Al Maktoum, Zayed University, United Arab Emirates
Reviewed by
Aan Hendrayana, Sultan Ageng Tirtayasa University, Indonesia
Nurul Istiq'Faroh, Surabaya State University, Indonesia
Updates
Copyright
© 2026 Alreiahy.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Sanaa Alreiahy sanaa.alreiahy@adu.ac.ae
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.