In recent years, governments have spent considerable time, effort and political capital on education reforms in the expectation that pupil attainment will improve. Since the mid-1990s reforms in England have focused on school governance issues, notably the weakening of local government control over schools via the academy system; greater pay flexibility for teachers – even within local authority controlled schools; reforms to the teacher training and recruitment systems; and pedagogical innovations in the classroom. Since Theresa May became Prime Minister there has been an increased focus on the value of grammar schools and selection more generally, issues that were last discussed seriously over thirty years ago.

England is not alone in seeking to improve pupil attainment and the performance of its schools. Education policy has been top of the agenda for many countries across the world: it is commonly accepted that a nation's competitive performance in a global economy reflects the skills and education of its workers which means they must either grow that talent from within or rely on high-skilled migrant labour.

Yet despite government efforts, academic and policy concern regarding pupil attainment in primary and secondary schools persists. These concerns reflect perceptions of high variance in educational attainment both across and within schools, even when accounting for differences in pupil intake. There is also some disappointment about the most recent results from the Programme for International Student Assessment (PISA) in 2015, which indicated little or no change in pupil performance in England since 2006 (Reference Jerrim and ShureJerrim and Shure, 2016). Geographical variance in attainment is growing, as indicated by PISA scores for Scotland and Wales, and there are concerns regarding regional variance in attainment within England with some pointing to a new North-South divide (Ofsted, 2016), while others point to a problem in coastal towns (Reference Ovenden-Hope and PassyOvenden-Hope and Passy, 2015). It is, perhaps, for this reason that parents have become more aware of the need to investigate the quality of education offered by competing schools.

So, what works in schools and how can governments bring in successful policies? Despite a growing literature investigating the determinants of pupil achievement, there remain substantial gaps in the evidence base in relation to how to improve both school and teacher effectiveness. For a number of years, governments have focused, perhaps rightly, on the quality of teacher training and the adequacy of conditions attracting able individuals into the profession. Even very recently the Department for Education has announced plans to provide further support and opportunities for new entrants to the profession (Department for Education, 2017). The importance of pedagogy has also been increasingly recognised, and recent years have seen substantial growth in the evidence base on interventions designed to improve pupil attainment, especially among the most disadvantaged, notably through much of the work funded by the Education Endowment Foundation. Increasingly though, the research focus has followed the policy debate in investigating the importance of school leadership and school governance, most notably in terms of the advent of academy schools and the growing autonomy that school leaders have over key decisions relating to the deployment of resources.

Reflecting this shift in emphasis, analyses are increasingly conducted at school-level. School-level analyses draw attention to the importance of an array of inputs that may affect educational outcomes for children that can be overlooked when focusing on the pupil–teacher interaction. These include the importance of non-teacher as well as teacher staffing; the mix of activities undertaken at a school and the time devoted to them; and teacher interactions, for example in terms of the mix of seniority in a school or within a team within a school (e.g. the maths teachers or English teachers). It draws attention to possible trade-offs headteachers and others face when devoting resources to one issue, sometimes at the expense of another. The emphasis on school leadership is akin to that on firm leadership in the Chief Executive Officer (CEO) literature – the way senior staff are paid, the autonomy they have, and how they are governed are considered at least as important as the interface between the teacher and the pupil.

This Special Issue contributes to the literature on school performance with four papers tackling the issue. The first paper by David Wilkinson and co-authors sets the scene by mapping the variance in pupil attainment in England's secondary schools over the period 2009/10 to 2015/16, and establishing how much of that variance is accounted for by the school a pupil attends. This is of first-order importance since, if schools account for little of the variance in attainment, there is arguably less need to be concerned about the role of schools in determining attainment. It is also important because, to our knowledge, this type of exercise has not been undertaken on a systematic basis since the early 2000s, and much has changed in the school system since that time. It might be reasonable to anticipate, for example, that increases in school autonomy through the introduction of the academy system may have led to greater variance in school performance.

It turns out that schools do matter – based on data for 2015/16, for example, they account for between 12.1 and 29.4 per cent of the variance depending on which pupil attainment metric one uses. The exercise is made extremely difficult by the way in which government moves the goal posts, setting then resetting the metrics by which pupil attainment and value-added are calculated. The authors find little evidence of a substantial change in the percentage of pupil variance accounted for by schools over the period; although it is a relatively short period it was, nevertheless, characterised by policy changes that are perhaps unprecedented since the advent of the comprehensive system in the 1960s. Most policy efforts are focused at school-level, even though variation across schools in pupil attainment remains fairly low.

The remaining papers in this special issue explore three aspects of schools that have been targeted by policy reforms: management practices, teacher recruitment (focusing on the use of newly qualified teachers), and subject choice.

School management practices have been a key area of focus for policymakers. The second paper by Alex Bryson and Francis Green examines the contention, which is at the heart of current government education policy, that state schools could benefit from the way private schools are managed. This idea rests on the perception of some government ministers that private schools' management practices are superior to those deployed in state schools, a belief that has, in part, led government to promote a school sponsorship arrangement whereby private schools ‘mentor’ state schools. Bryson and Green examine the prevalence of a range of human resource management (HRM) practices in state and private schools. The idea that HRM practices matter for organisational performance is broadly accepted in the management and economic literatures, but this paper is the first to examine the use of HRM in private and state schools and consider their implications for an array of school performance outcomes. The authors find that, contrary to the government's position, HRM is used more extensively in state schools than in private schools, and that this is the case when comparing observationally similar schools. Furthermore, these practices are only positively associated with improvements in school performance in the state sector. The authors look at a broad range of performance measures including financial performance, labour productivity, and quality of product/service, although they are unable to look specifically at attainment, which would be a worthwhile topic for future research. These findings raise potentially important questions about the direction of current government policy. Of course, there may be other forms of management and leadership practices outside of those considered by the study that may be beneficial for performance. It may also be the case that there are still valuable lessons to be learned from the highest-performing private schools.

The quality of teaching has long been identified as a critical determinant of pupil attainment, hence the focus of many reforms on teacher recruitment and training. The third paper by Sam Sims and Rebecca Allen explores the use of newly qualified teachers (NQTs) by schools in England. Recent years have seen declining rates of retention among NQTs, and Sims and Allen use a novel method, typically applied in the health statistics literature, to identify schools that both hire an unusually high level of NQTs, and also those which lose more NQTs than would be expected. While this group of schools is relatively small overall, they account for a disproportionate share of attrition from the profession. The authors suggest that such schools are likely to be those with poor working conditions, and that NQTs are often not well placed to identify such schools in advance of accepting a post. The authors propose that their method may help provide this information to NQTs, and at the same time provide an incentive to schools to improve their working conditions and attractiveness to staff. Implementing such an approach would however require careful consideration so as not adversely to impact on schools which may already be operating in difficult circumstances and struggling to recruit. Such schools are also likely to require support in implementing change so that they become better places for NQTs to work.

The final paper in this issue, by Jake Anders and colleagues, considers the role of subject choice. The curriculum that pupils study from ages 14–16 formed a key part of the 2010–15 UK government's education reforms. This included the introduction of the English Baccalaureate (EBacc) performance measure for schools, incentivising study of “subjects the Russell Group identifies as key for university study” (Reference GibbGibb, 2011). Anders and colleagues investigate whether young people who study specific sets of subjects to age 16 have different probabilities of entering university, and specifically, a high-status university. The authors find that conditional differences in university entry attributable to subject choice are, at most, small. However, as the authors acknowledge, this does not mean that subject choice should be seen as unimportant – there may be important consequences, for example, for the eventual subject studied at university, even if effects are small when considered solely in terms of whether the individual attends university or not.

So what can we learn from all this? While schools may appear to account for a relatively small part of the variance in pupil attainment, their contribution is certainly not insignificant. And as far as we can tell from the available data, their contribution also appears to have shown little change over time. So schools do matter, although it is important not to lose sight of the fact that other factors outside of the school system also play a substantive part in determining young people's outcomes.

One key lesson is the difficulty in assessing school performance. While educational attainment is of considerable importance, even attainment measures are not as straightforward to interpret as one might first think, with results often sensitive to the measure adopted. Arguably, more attention should be paid by academics and policy analysts to the production of standard metrics for assessing pupil attainment over time, nothwithstanding the usefulness of creating measures that best capture school quality, such as contextual value added measures. There is also perhaps a trade-off between more complex measures that may better capture school quality and those which are more readily understood, but which do not reflect pupil intake.

It is also important to remember that attainment is not the only metric by which we might judge school performance. Understandably, much of the literature on schools focuses on attainment – and notwithstanding the above comments about changes in metrics, data on attainment are readily available, and we are fortunate in England in having access to large-scale administrative datasets that allow us to undertake such studies. But we should not forget that other outcomes of schools are important too – the non-cognitive skills that pupils develop are also likely to be an important factor in determining their outcomes in later life. With advances in administrative data and developments such as the Longitudinal Education Outcomes (LEO) data, in future it should prove possible to look beyond attainment and at longer-term impacts of schools on earnings and employment, which represents an exciting opportunity for future research.

Even though other factors are likely to dominate in determining young people's chances of success in life, arguably schools and their staff are areas in which it is more feasible for policy to achieve real change. Wilkinson et al. show that much of the school-level variance remains unexplained after accounting for school composition, which suggests further exploration of the factors responsible is warranted. The remaining papers in this issue have explored some key areas that may provide answers: management practices, the role of newly qualified teachers (and the working conditions that can support them), and the curriculum studied. The findings from these papers suggest that all of these factors are likely to have a role to play in improving school performance. Unsurprisingly, attempts to raise the effectiveness of schools are likely to require various strategies, with no silver bullet solutions. But continuing to extend the evidence base on what works in schools is of clear importance. Only then perhaps, can we truly work towards achieving better schools for all.