School leadership: Time to end the hero head myth

Written by: James Croft | Published:
Image: Adobe Stock

With a significant proportion of headteachers approaching retirement, a report has raised concerns that the government’s leadership succession plan is still too reliant on ‘hero heads’. Its author James Croft says it is time to think differently about leadership and give schools more responsibility and trust

In an increasingly autonomous schools landscape, much of the government’s education reform strategy has come to turn on its being able to capitalise a leadership premium.

With a significant proportion of heads approaching retirement, policy-makers have become acutely aware that we lack a strategy for identifying and developing leaders. Despite a massive literature that has built up around the subject, we in fact know very little about what makes for effective leadership, how it interacts with other factors of importance to school improvement.

A common assumption

A key assumption – prevalent in the literature, and implicit in much talk of the importance of leadership in education – is of leadership’s direct impact on academic outcomes. This idea has obvious appeal – we want to believe that whatever challenges a school faces, overcoming them is really just about finding the right person for the job – but of course it is much more complicated than that.

Few studies have tried to actually quantify the direct contribution of leadership, but those that have find leadership variables are only modestly to weakly related to pupil outcomes.

This shouldn’t surprise us – it stands to reason that whatever influence a leader may have is mediated through a number of school-related factors that are more proximal to the student level – most obviously teachers themselves.

Unfortunately, this doesn’t appear to have checked enthusiasm for theories of leadership that big-up the leader as the sole cause of change in organisational performance. The truth is that, even where strongly charismatic personalities are involved, failure is highly likely if organisational “fit” is lacking – as witnessed by the mixed results of successive government “super head”-type initiatives.

Leaders’ influence

A consequence of the dominance of trait and behaviourist approaches to identifying and developing leaders has been that a much more important set of questions, relating to precisely how leaders exercise their influence, has been overlooked.

Leaders evidently shape the conditions under which teachers work and their decisions are key for motivating them and focusing their efforts. What decisions and practices are important for productivity?

How is the influence of leadership enabled or compromised by contextual factors? How might stakeholder responses – even pupil outcomes themselves – in turn influence leadership decisions and behaviour?

Nevertheless, an increasing minority of quantitative studies now provide some direction with respect to the most important means by which school leaders bring their influence to bear on organisational effectiveness.

These include mission and goal-setting, variables related to the setting of the curriculum, and the provision of instructional guidance for teachers.

The importance of these mediators is corroborated in the economic literature by a number of studies of “school autonomy” – the result of reforms to governance often alternatively referred to as “decentralised decision-making” or “school-based management”.

In addition to changes to leadership and management structure, this research indicates that the scope to shape curriculum and instructional method, and the motivation of staff (specifically through the use of appropriate pay and conditions incentives), are important for raising academic achievement too.

These are useful, if broad, pointers – but they only take us so far. The leading question for social scientists is, after all, not “what works?” but “what works for whom, how, and under what circumstances?”.

To get at these questions we need a model, and methods, that do justice to the complexities of heads’ relationships and interactions with staff, pupils, parents, and others, in context. Unfortunately, work to this end has barely begun: there’s been progress with modelling these relationships, but prevalent research strategies in the field are simply not up to the job of testing them.

So what do we do? We cannot wait on research that may never materialise: we have schools to run. A number of policy implications follow.

No more heroes

First, we need to stop looking to “hero heads” for all the answers and desist from building policy and national school improvement strategy on the basis of exceptions to the rule. There are too many leadership “types” that feed this kind of thinking, and it doesn’t help us.

The reality is that we know little about what particular practices are impactful, learnable and transferable. This being the case we should not assume, for example, that independent school heads have all the answers for turning around underperforming state schools, or that grammar school heads, by virtue of their success with bright kids, have the leadership capacity to devise teaching and learning strategies for non-selective school environments, as presently mooted in some policy circles.

Professional qualifications

Second, and for the same reason, it’s right that participation in leadership development programmes should no longer be required for service.

Leaving aside that we do not have even a vaguely meaningful proxy for quality of headship, there is no real evidence to support claims that professional qualifications make a difference.

This is perhaps unsurprising if, in effect, everybody takes the qualification and everyone passes – but even if we are more selective at the point of entry, it’s unlikely this will make much difference unless course content is improved.

Certainly many elements of the courses on offer address aspects of the leadership brief that aspiring leaders can’t afford to overlook – “how to” aspects of preparation for headship like financial management, understanding inspection requirements, setting pay and conditions, and performance management, to name but a few – but there’s also a lot of discredited (though nonetheless persistent) theory on “leadership style” and qualities.

The problem is that national professional qualifications by definition must be comprehensive; they’re designed to leave leaders fully equipped to meet the challenges of any context. They don’t tend to discriminate terribly effectively between the necessary and the important – let alone the superfluous. Such courses rarely make allowances for prior experience.

The really important stuff may equally well be acquired on the job, by less formal means, online or in modular fashion, as required. But research producer interests tend to dictate an all-inclusive programme offering. There’s little appraisal of the evidence-base for the specifics of what is typically required for such qualifications; debate about their value is typically ended with the assertion that (merely having) a national, professional qualification for headship (in and of itself) raises the profile and status of school leadership.

So what should we do?

Given the contingent and evolving nature of leader emergence, and the individual and contextual nature of factors impinging on recruitment and retention decisions, it follows that the locus of leadership identification and development should be shifted to the schools level.

If school leaders themselves cannot identify potential leaders, and design and develop opportunities for leadership development in school, there is certainly no basis for believing this to be within the skill-set of central government.

There are capacity issues to consider with this approach, to be sure. It seems likely that federations and academy chains may be better able to match support with opportunities for career progression on this model.

Accordingly, it would make sense for all Teaching School Alliances to be set up within multi-academy trusts. But barring the more focused taught component outlined, leadership development should be provided by leaders of similar schools, in situ. Supporting potential and emerging leaders should be a regular and routine whole-school activity.

Likewise, peer-to-peer support for already practising leaders should be supplied according to demand from those in need of advice, strategy, and support, in the leader’s context. This gets around the misleading sense that an intervening “hero head” is in some way taking responsibility for the outcomes of the advice he/she has given – at present a significant flaw in the practice of collaborative approaches to system improvement.

Ultimately what’s needed policy-wise – the framework within which we should be discussing what being qualified for headship looks like – is a clear commitment to research literacy. Including an elective module in the present NPQH on “using data and evidence to improve performance” is not nearly adequate.

Those being trained into (or developed in) leadership need to be continually building and refreshing knowledge of what has been shown to work generally, in specific relevant contexts, and for different profiles of student. It is clear that, in consequence of the increasingly outcomes-focused direction of performance evaluation, a significant change in the way practitioners interact with the research literature is already underway. For this to become embedded, the inspectorate too must grasp the importance of validating its judgements with reference to evidence of what works.

The government, however, seems at present to be set on re-investing in hero head-based strategy and a suite of leadership development qualifications to be provided by the National College for Teaching and Leadership.

This is in accord with an increasingly explicit preference for centralised decision-making – as witnessed by its on-going control of the recruitment of academy sponsors, and increasing centralisation of decision-making in respect of national educational standards and assessments, regulation, and accountability.
It hardly needs pointing out that this tendency ultimately only acts to subvert the purpose of autonomy reforms by taking over leadership recruitment at the top tier and reducing leaders’ scope to take decisions and effect strategies in the areas identified to have bearing on academic improvement.

In that the potential impact leaders can have on organisational effectiveness increases with the degree of decision-making autonomy afforded them, the government would do well to be more modest in its ambitions.

  • James Croft is the executive director of the Centre for the Study of Market Reform of Education and author of its recent report Taking a Lead: How to access the leadership premium.

Further information

Taking a Lead: How to access the leadership premium can be downloaded at


Please view our Terms and Conditions before leaving a comment.

Change the CAPTCHA codeSpeak the CAPTCHA code
Sign up SecEd Bulletin