Is there a place for place in educational attainment policy?

ABSTRACT This paper aims to examine the case for a focus on place-based drivers of inequalities in educational attainment among secondary school students in Scotland. Using desk-based sources, it provides an account of the post-2015 policy episode around improving educational attainment among children from disadvantaged areas. This started with the Scottish Government claiming that its ‘defining mission’ was to ‘close the gap’ but the place-based focus of policy was soon dissipated and the legislation that intended to be the flagship of reform was shelved. The paper shows that international evidence prompts a need for the impact of disadvantage based on place to be factored into approaches to schooling and provides a regretful account of its insecure traction in Scottish policy. It argues that a serious weakness of the case for place in Scotland is the underlying evidence base and concludes by suggesting how existing data sources could be used more effectively and by outlining some alternative policy approaches.


Introduction
Ahead of the 2016 Scottish General Election, there was a stirring within the educational policy community in Scotland, prompted by Nicola Sturgeon as First Minister as she made a series of speeches about educational inequalities, highlighting that the school careers of children from disadvantaged areas too often culminated in lower attainment and unsatisfactory post-school destinations. The manifesto of the Scottish National Party (SNP) for the 2016 Scottish general election then introduced a new flagship policy in the following terms: 'Ensuring educational excellence for all and closing the gap in attainment between young people from our most and least deprived communities will be the defining mission of the SNP in the next parliament' (Scottish National Party, 2016, p. 8).
The re-election of the SNP to government was followed by immediate announcement of actions, funding and forthcoming legislation. John Swinney, Deputy First Minister and former Cabinet Secretary for Finance, was appointed to the education brief, which was widely viewed as confirmation that educational attainment -and reform -was right at the top of the new Scottish Government's priorities (e.g. BBC News, 2016).
The emergence of attention to place-based inequalities was remarkable because the debate on this topic had been subdued for many years. But the promise that the incoming government would overturn the neglect of place as a key dimension of educational inequality was to be short-lived. As soon as substantive policy papers began to be published it became clear that the focus on place was insecure, and within two years the Education Bill that was to lead the charge to 'close the gap' had been shelved.
This paper aims to examine the case for 'a place for place' in educational attainment policy in Scotland (in other words, the arguments for focussing on place as a key dimension of educational inequality) and to explore why it failed to gain traction. It first presents some background to the emergence of concern about place-based disadvantage in schooling in Scotland, whose policy environment is distinctive within the UK. Second, from the international literature, it considers why place is important to schooling, especially thinking about the ways in which place and schooling are mutually constitutive. Third, it examines the policy episode around closing the gap, especially the fading of the importance of place as a key aspect of inequality. The paper closes by making a case for a new intelligence strategy to better understand place-based inequalities in Scotland in order to better inform future policy and provides some pointers as to what such a policy might look like.
The paper was prepared using desk-based resources in a four-part strategy. The background on Scottish education policy is based on a reading of the educational administration literature on Scotland, starting with standard works and widening out to literature that has considered the persistence of inequalities. The material on place and its links to educational disadvantage is a development of literature reviews conducted by Lupton and Kintrea (2011) and Kintrea et al. (2015). It is influenced by understandings drawn from economics and sociology and developed within the field of urban studies that place can be a source of disadvantage and not only a reflection of it, summarised, for example, by Galster (2019). The review of the Scottish Government's approach to 'closing the gap' is based on a comprehensive reading of policy documents published between 2015 and 2018, interrogated particularly for references to place-based inequalities. This material was supplemented by published commentary, media reports, proceedings of the Scottish Parliament Education and Skills Committee, and some later policy material.
Disadvantage in educational outcomes is evident among pre-school children and continues (and widens) through school careers and into adult outcomes beyond (e.g. Goodman & Gregg, 2010;Sosu & Ellis, 2014), which is fully recognised by the Scottish Government. However, the focus of the paper is principally on secondary education, largely because it has been the focus of much of the post-2015 debate, even if that focus has itself been shaped by inadequate data availability, which is considered later in this article.

A distinctive policy context
Compared with the conduct of education reform in England by recent UK Governments, the Scottish environment has been altogether much calmer. However, schooling in Scotland is influenced by a similar high-level policy agenda, with recognition of the importance of education to the globalised knowledge economy, as well as concerns about social justice. Scotland's adherence to this agenda became stronger after the publication of an OECD report (Teese et al., 2007) on the quality of Scottish schooling (Audit Scotland, 2014;Mowat, 2018). Some argue that English and Scottish education systems have both become fully neoliberal. Poole and Mooney (2006), for example, claim that there is a drive to 'nurturing a British edubusiness sector' (p. 582). However, the school system in Scotland contains far fewer market elements. Scotland's comprehensive secondary school landscape emerged during the 1960s and has carried through into the 21 st century (Paterson, 2003). 'State schools' are non-selective and co-educational and managed by local authorities, which also employ their staff and control funding. Local authority management extends to 'denominational' schools, which at secondary level are all Roman Catholic.
A proposal to introduce routine testing of children was rejected in the 1980s and not implemented in Scotland (Humes & Bryce, 2013), and testing continues to face resistance. Therefore, the basis of school 'league tables' is restricted to published results of secondary school examinations. The government itself does not publish exam results by school, although they are an annual feature in the Scottish media.
Place is a strong feature of school organisation. By law, local authorities must define school catchment areas and the default is for children to attend the school within whose catchment they live. Quasi-markets as an influence on access to schools appear to be relatively weak compared with England. There are elements of parental choice in Scotland but there is tight national guidance about how it can be exercised, and most children attend their local school.
In terms of international measures of attainment, Scotland's schooling appears to be moderately successful. In 2012, it lay 18 th out of 37 countries in the PISA ranking, 1 i.e. above average but well below the highest performing countries (Boyling et al., 2013). However, in 2015 the PISA scores for Scotland fell for mathematics, reading and science (Organisation for Economic Co-operation and Development [OECD], 2016; Scottish Government, 2016a) with reading scores improving in the 2018 PISA, but mathematics and science continuing to fall (Scottish Government, 2019a). Performance in mathematics and science is now similar to OECD averages with reading somewhat above. Whatever the reason for these changes, observations that Scotland's performance sits within international norms (OECD, 2015;Teese et al., 2007) still remain convincing. 2 Overall then, Scotland's secondary schooling seems relatively uncontroversial. Given its 'comprehensive universalism' it might be imagined that there would also be common experience among young people in obtaining benefits from education, and thereby access to higher education and the labour market, or at least benefitting according to their ability. Yet the evidence is of a strong gap in educational outcomes between young people from different backgrounds, as both Croxford (2015) and Mowat (2018) show. Beneath the calm surface of Scotland's consensual and seemingly inclusive approach lie significant inequalities.

The role of place in the attainment gap
In examining Scotland's attainment gap, there is an important question of what the gap is a gap between. In a school system, there may be an attainment gap between students from different socio-economic backgrounds, and a gap between young people who attend different schools. There may also be a gap between young people who live in different places or neighbourhoods. In trying for more precision it does not help that these three gaps intersect. Although this paper's concerns are about socio-economic disadvantage in particular places and how it plays out, there are many other potential intersecting attainment gaps. For example, gaps may relate to gender, to ethnicity, to migration status, to those with experiences 'in care' as well as those with additional support needs and disabilities (see, e.g. Strand, 2014). This paper uses 'place' as a shorthand for the combination of contextual factors that are experienced differently according to where people live. After Samson (e.g. 2019), I assume that where people live is important in shaping their life chances in its own right and can also mediate both wider structural forces and individual processes, such as choice. Riley (2013Riley ( , 2017 shows that places, representing a set of physical, socio-economic, cultural and relational contexts are inextricably linked to the lives and experiences of children and young people across the globe, therefore she argues they should be central to thinking about educational practice. This is especially relevant in an era when urban habitats across the world have become segregated as never before (Nightingale, 2012). Place has no precise scale but it suggests a level at which a sense of commonality and local shared experience can be found (Altman & Low, 1992). Many academic writings on place, including some drawn upon in this paper, use the alternative word 'neighbourhood', which tends to indicate a relatively small, identifiable residential area with common features (see, e.g. Kearns & Parkinson, 2001).
Place can be a slippery concept in education (Nespor, 2008) and language in the discourse in Scotland is often inconsistent, with reference often made to 'deprived areas' or 'deprived communities'. 'Community' is a long-established alternative way to refer to place in public policy, with a degree of rhetorical warmth and an inbuilt connotation of people and place together. There is no doubt, however, that the initial focus of the Scottish Government was specifically about closing the gap in attainment between places, understood as residential areas with different characteristics. The First Minister personally initiated the attainment gap policy and made numerous contributions in the run-up to the 2016 Scottish General Election. She referred explicitly to differences in attainment between young people from deprived and non-deprived areas as one of the biggest challenge of Scottish education: 'too many children still have their life chances influenced more by where they live, than by how talented they are, or how hard they work' (Scottish Government, 2015a). She reinforced that she was referring to place by mention of her own school experiences in her home town.
News reports and comment pieces show that the understanding of journalists and academics alike was that the attainment gap that the government proposed to attack was the gap between places. For example, Hepburn (2015) cites academics from two universities whose interpretations of the First Minister's announcement was that its focus was on deprived places (for further evidence of this understanding see Cramb, 2015;McKenna, 2015aMcKenna, , 2015bMcNab, 2015). In the SNP's election manifesto, it was the gap in attainment between 'communities' that was highlighted (Scottish National Party, 2016) but the context made it clear that the word was intended to denote place-based communities in particular.
As its actions for policy change proceeded, the Scottish Government used the Scottish Index of Multiple Deprivation (SIMD) (Scottish Government, 2016b) as its measure of disadvantage and its means of targeting earmarked funding. 3 The SIMD is a broad socioeconomic measure intended to capture diverse aspects of the absence of well-being as well as elements of low income, calculated for small areas across Scotland. As with all similar indices, an area's low rank in the SIMD does not in itself mean that all the people who live there are personally experiencing deprivation. Conversely, many individuals and households who experience deprivation do not live in SIMD-defined deprived areas (see McKendrick, 2018). Ideally, data would exist to unscramble socio-economic and institutional disadvantage from place-based disadvantage as it relates to educational attainment. Nevertheless, the SIMD provides an accepted measure of the extent to which people living in different places experience a lack of well-being. Its use as the primary measure of disadvantage in the attainment gap project reinforces the understanding that the policy focus was on places.
The government's priority followed an emerging concern about differences in attainment between places. Older studies in Scotland, such as by Garner and Raudenbush (1991) and Pacione (1997), had shown that neighbourhood socioeconomic status had an impact on educational attainment among secondary school students. In the current era, a wider understanding has emerged of the association between levels of attainment and neighbourhoods at different points on the deprivation rank particularly from the work of Sosu and Ellis (2014), although not of clear causality. They used a variety of administrative data and showed that gaps widened between primary and secondary levels, and continued beyond school into further and higher education, prompting a wide debate. 4

'The spatial school system' and neighbourhood effects
International research shows that place-related attainment gaps emerge from a series of interlocking, dynamic factors that derive from the interaction of local agency, social contexts and public policy that together create a 'spatial school system'. As Thomson (2002) observes, schools are context-derived but they are also context-generative (p. 73).
At their simplest, place-based inequalities in schooling derive from the relationship between socio-spatial segregation and locally based catchment areas. Where there is strong residential segregation and the default is for children to attend their local school, intakes typically demonstrate distinct socio-economic characteristics. These factors are particularly strong in urban Scotland: there are distinctive, historic patterns of urban development arising from industrialisation and de-industrialisation, and from public policy towards housing development (Walsh et al., 2016). This means that an important part of the context for schooling in Scotland is very deep and longstanding socio-spatial segregation, leading to a narrowing of the social mix in individual schools (Murphy, 2014).
Since 1980, the impact of residential divisions on schools in Scotland has been accentuated in two ways. First, the introduction of even limited parental choice within a relatively flat school landscape has created elements of a quasi-market, whereby some parents seek to gain access to the 'better schools'. Older studies (e.g. Echols & Willms, 1995;Willms & Echols, 1992) found that parents who exercised placing requests tended to be more highly educated and in better jobs. Willms (1997) concluded that segregation between schools had increased since choice was introduced. By 2010, secondary school placing requests had risen to about 14% of the school roll nationally and to over 25% in five urban local authority areas (Scottish Government, 2010). No more data have since been published but there is no reason to suggest that the importance of placing requests has diminished.
The second dynamic factor concerns how schools interact with the housing system. Housing markets internationally exhibit premium prices for access to the 'best schools' (e.g. Croft, 2004;Machin, 2011). The price differential then further serves to accentuate the association between places and social class (Glen & Nellis, 2010). By the 1990s, the majority of households in Scotland had become homeowners (Gibb, 2015), enabling greater numbers of households to exercise residential choice. Although there have been no studies in Scotland, catchment areas are routinely highlighted by estate agents, and informal evidence from sources such as the Mumsnet website suggests a pattern of house moves by better-off families to favoured secondary catchments.
The wider literature also points to disadvantaged contexts making it more difficult to provide good education, which is the third factor. Schools are context-derived in the sense that their everyday operations are affected by their embeddedness in particular places. This includes the influence of a preponderance of families that have a lack of income to pay for extra-curricular activities and with whom schools find it harder to engage (Mowat, 2020). Research shows that a local concentration of students from disadvantaged backgrounds exerts a downward pressure on the quality of school experiences (Levin, 2007), partly because children suffer more insecurity and stress (Archer et al., 2010) and have less sense of belonging to the school community (OECD, 2017). Leaders in such schools must spend a greater proportion of their time supporting disadvantaged students and handling pastoral, attendance and disciplinary matters (Gewirtz, 1998;Lupton, 2006), as preconditions of improving attainment.
Teaching quality is also related to place composition. Bell (2003) and Lupton (2005) both found that poor neighbourhoods were associated with poor teaching quality, including difficulties in attracting qualified staff. Thrupp and Lupton comment that 'low expectations and unchallenging work were in evidence in all the (disadvantaged) schools we studied' (2010, p. 318). Lupton (2006) noted that the regulator's ratings of school quality aligned with the index of multiple deprivation, which is also confirmed by more recent research (Clifton & Cook, 2012, and see;Roberts, 2018).
The fourth dimension of the impacts of place on education is the potential for 'neighbourhood effects'. The question is whether disadvantaged neighbourhoods merely cluster disadvantaged people, or if they also embed their residents in a context that further activates disadvantage (see, e.g. Kerr et al., 2014;Van Ham et al., 2013). A key theoretical mechanism is collective socialisation, effected variously through role models and peers at different levels of deprivation (e.g. Andersson et al., 2007;Galster, 2007). Recent work suggests the existence of complex, cumulative and path-dependent loops that operate between individuals and across generations (Galster & Sharkey, 2017). Neighbourhood effects research suggests that residents in disadvantaged places will adopt negative or perhaps even fatalistic dispositions towards education because they have developed identities that are shaped by their local experiences (Atkinson & Kintrea, 2001;Quane & Wilson, 2012).
There have been many studies that have sought to quantify neighbourhood effects on educational attainment (see Bower & Rossi, 2019;Lupton & Kintrea, 2011;Nieuwenhuis & Hooimeijer, 2016). Nieuwenhuis and Hooimeijer conducted a meta-analysis of 88 studies in developed countries and concluded that neighbourhood poverty has an independent influence on educational attainment. They raise the question as to how specific neighbourhood influences on educational attainment can be disentangled from other related factors such as parents' socio-economic status and school factors, but their conclusion is that neighbourhood has an independent effect after controlling for such influences.
Altogether, the literature shows that place matters for schooling because there is a recursive relationship between schools and places: the 'spatial school system'. Local social composition, which ultimately has its source in the economic structure, shapes school intakes and affects the housing market, which in turn further influences social composition. Further, the composition of pupils is typically shaped by defined catchment areas and accentuated by quasi-markets and it can affect the management of the school and educational processes, and therefore the quality of education. Finally, the social composition of a place potentially shapes the world views of residents. In consequence, in disadvantaged areas, there may be a risk of negative attitudes towards school and low aspirations that diminish educational attainment.
Empirical research on these themes in Scotland is underdeveloped, therefore caution must prevail. Perhaps the particularities of its universal comprehensive system provide some mitigation of place-based influences. It would also be a mistake to elevate place as a driver of inequality in attainment above the other key dimensions discussed above. But the evidence from elsewhere suggests that the spatial school system, which feeds both from and into the social composition of schools, is likely to have important effects on teaching and learning processes in Scotland, particularly given the strong socio-spatial divisions that prevail in its urban areas.

The short life of a policy episode
With evidence of a wide attainment gap between young people from less and more disadvantaged areas and amid broader fear that attainment was falling, the 'closing the gap' initiative by the 2016 SNP government was a very welcome development. Education Scotland's Corporate Plan had previewed a direction change and highlighted differences in attainment by SIMD rank as 'a key issue holding us back' (Education Scotland, 2013, p. 19). 5 Not long after, the Scottish Parliament's Education and Culture Committee held an enquiry into the attainment gap. However, the committee focused on disadvantaged households and chose not to examine place-based inequalities, instead focussing on the potential roles of teachers, schools, parents and employers (see Constance, 2015;Maxwell, 2015). This was already suggestive of a certain insecurity of grip by policymakers on the place-based aspects of educational inequalities.
The government's first action on place-based inequalities was to launch the Attainment Scotland Fund (ASF) in 2015 to provide extra resources to schools in areas of high deprivation in nine local authority areas. Its aim was stated as 'to help achieve equity and address the priority to close the attainment gap between children and young people living in our most and least deprived communities' (Education Scotland, 2016). Money goes directly to schools, with a significant emphasis on improving learning and teaching and supporting children with particular needs. However, from the outset there appeared to be a lack of clear thinking about the distinction between place-based, socioeconomic and institutional drivers of inequalities, and no sense of the potential role of 'spatial school system'.
Further initiatives quickly followed, including an education summit, a funding initiative to support head teachers, a new international council of education advisors, and a delivery plan that promised 'a relentless focus on closing the attainment gap' (Scottish Government, 2016c, p. 4). A statement by the education secretary to the Scottish Parliament's Education and Skills Committee provided assurance that the focus was on place: 'we want to ensure educational excellence for all by closing the gap between young people from our most and least deprived communities' (Swinney, 2016, p. 1). But the subsequent introduction of the Pupil Equity Fund (PEF) (Scottish Government, 2017b) suggested that the government's focus on attainment was not all about places. PEF allocates a pot of money to schools based on the proportion of children on free school meals, which is a marker of household rather than a place-based disadvantage.
Funding going directly to schools was an indication that shifting power away from local authorities was also an objective. Following a discussion paper (Scottish Government, 2017c), the Scottish Government consulted on an Education Bill that would establish a 'head teachers' charter', intended to grant more autonomy over the curriculum, improvement plans, staffing and finances (Scottish Government, 2017d). It also proposed to establish six statutory 'regional improvement collaboratives'. However, there was no articulation of the theory of change that linked the empowerment of head teachers to reducing place-based inequalities, so this appeared to be a parallel agenda. 6 After several further discussion papers and consultations, it became clear by mid-2018 that the government could not proceed with the bill, which it had described previously as 'the centrepiece of the legislative programme for the year ahead' (Scottish Government, 2017h). This was because it faced opposition from local authorities and, as a minority government, from other political parties, especially on the head teachers' charter. A draft bill (Scottish Government, 2018a) was published but not presented to Parliament, so it was effectively shelved. Instead, the government concluded an agreement with the Convention of Scottish Local Authorities in which both signatories promised to continue with other elements of the erstwhile bill (Scottish Government, 2018b). There were no references to place-based inequalities either in the bill or the agreement.
Since then, the ASF and the PEF have continued, representing a £750 m commitment to 2021, supported by the (non-statutory) regional improvement collaboratives and attainment advisors, all tasked with closing the now broadly specified 'poverty-related attainment gap'. The sense that the place focus is now diluted is reinforced through an examination of the work of the regional improvement collaboratives. For example, the vision that frames the West Partnership Improvement Plan covering the Glasgow city region, wherein lie some of the biggest disparities in attainment between places, makes no specific mention of place (Glasgow City Region Education Committee, 2019, p. 7). And while its action plans include one about 'families and communities' there is no sense that the mutually constitutive relationship between schooling and places is on the agenda.

The problem of evidence
The closing the gap project also served to confront the government with important weaknesses in the evidence base for policy. The literature points to socio-spatial disadvantage combined with the 'spatial school system' as key factors underlying the attainment gap. There is, therefore, a need for a strong understanding of the interface between socio-economic factors, schools and place. When the debate was starting, a journalist referred to 'Scotland's educational apartheid' (McKenna, 2015a) but that analogy might have been more apposite if division between places had been more conspicuous. There is unfortunately very little precise understanding of the contemporary geography of attainment in Scotland, nor is the balance of place-based, socio-economic and institutional factors in its make-up known. Therefore, there is a weak basis for understanding how interventions might close attainment gaps.
After Education Scotland highlighted the attainment gap, it took four years for the Scottish Government to publish a consultation document on measuring it (Scottish Government, 2017f), with subsequent confirmation in the National Improvement Framework (Scottish Government, 2017a). The government settled on its definition of the most and least disadvantaged children and young people as those living in the bottom and top 20% of areas by SIMD rank (i.e. a place-based measure). However, the implications of choosing such broad groups for comparison was not discussed, even though its own evidence showed a strong gradient of attainment within the bottom quintile (Scottish Government, 2017e). The main indicators comprise literacy and numeracy measures at different age points, the achievement of one or more school qualifications at different levels of study, and a measure of the percentage of 16-19 year olds participating in education, training or work. 'Stretch aims' for each indicator also have been published, specifying the level of improvement desired in future years. They apply to all quintiles of the SIMD, not just the most disadvantaged areas. Effectively the stretch aims look to lessen the gradient of attainment between pupils from more and less deprived places and to raise the level of attainment of pupils across all SIMD ranks. These indicators are now in use although so far attainment outcomes are little changed (Scottish Government, 2019d). But in order to understand the attainment gap and take effective action, it is important to unscramble the factors that underlie it. They include both the separate and the conjoined influences of students' household background, the schools they attend, as well as the places where they live. In this article, the place-based elements have been accentuated because they appear to be potentially very important in the context of Scottish urban settlements that have strong socio-spatial divisions, and have been neglected (and poorly understood) in the Scottish context.
Currently, little is known about the attainment of pupils from different household backgrounds. The government trailed 'a bespoke index of social background' containing individual-level data (Scottish Government, 2017f) but no further announcements have followed. Although there exists an annual parents' data check, this does not collect data on incomes, occupations, ethnicity or parents' own education, and there is no equivalent of the National Pupil Database for England.
Concerning schools, public data is limited to their examination results (therefore to the 'senior phase' of secondary schools) and to inspection reports. As part of the post-2016 reforms the Scottish National Standardised Assessment (SNSA) was introduced for pupils in primary years 1, 4, 7 and secondary year 3 (Scottish Government, 2017g). Although this has the potential to provide new attainment data at school level across a range of age groups, it is controversial, with a particular furore around tests for 4 and 5 year olds (e.g. Ross, 2018) that prompted the government to promise modifications (Scottish Government, 2019b). But more importantly for the concerns of this paper, the Government has agreed that most data will not leave schools and that government will have access only to national-level data. The intention is to avoid school league tables but it seems remiss if the opportunity has been forgone to carry out any school-level analysis. 7 There are other existing data sets about schools that could provide important insights into how the spatial school system works. Scottish Government conducts a regular teacher census, which includes the age, gender, working status, ethnicity, employment status, and subject specialisms of teachers, and from which could be calculated the turnover of teachers in particular schools. However, at present these data are not brought into the debate in order to explore the relationships between the characteristics of schools that pupils attend and their attainment.
Concerning place-based factors, there would also be an advantage in documenting the precise geography of deprivation in relation to schools. Existing data show the proportion of young people in secondary schools who have addresses within each of the SIMD quintiles, but it is not clear whether the results are mainly a consequence of the design of catchment areas or whether they are significantly influenced by parental agency. A study using suitably anonymised individual pupil attainment data, held by the Scottish Qualifications Authority -and ideally also from the SNSA -would help to penetrate this gap.
A final part of understanding what underlies the attainment gap lies with the dynamics of the spatial school system. There has been no public information for 10 years now on how many placing requests are made, nor how many are granted, nor their geography. Given all the evidence of the use of 'choice' by more advantaged parents to secure further advantages for their children, this is also an important missed opportunity.

So is there a place for place?
Looking across this episode, 'closing the gap' was explicitly launched as a means to tackle place-based inequalities in attainment. But it is apparent that the clear focus on place that had been articulated by the First Minister and her education secretary was much less secure than statements suggested. The reasons for this require deeper investigation but there were certainly fully legitimate concerns that other dimensions of inequality in attainment were also important, evident in discussions in Parliament (Scottish Parliament Education andSkills Committee, 2016, 2018) and reflected, for example, in the PEF. Mowat (2019b) shows that the debate about social justice in Scottish education is complex, with competing discourses about international competitiveness and different conceptions of social justice entering the fray. It was unfortunate that the government's mission became embroiled in controversy over the head teachers' charter, whose contribution to closing the gap could have been better articulated. The government was also coming under more pressure to improve pupil attainment as a whole, as PISA and other evidence sources continued to fuel criticism. Throughout, key issues surrounding the origin and persistence of the place-based attainment gap were subdued within the policy discourse, confounded by a weak evidence base. There was also little or no apparent understanding of what this paper calls 'the spatial school system'. Policy implementation was all about what could be done in schools, rather than also about the community-or area-based approaches that might have been a logical approach to place-based disadvantage. The isolation of schooling from the agenda on places is somewhat surprising given a record of more than 30 years of urban regeneration in Scotland (Robertson, 2014) and ongoing policy towards improving disadvantaged places, shaped by the Community Empowerment (Scotland) Act, 2015.
The question arises, then, about what a place-based element of attainment policy might look like. The purpose here is not to propose specific actions but to highlight pointers from other work that indicate four possible directions. First, encouraging socially mixed neighbourhoods is central to addressing the socio-spatial segregation that shapes school intakes (Robison et al., 2016). Policy interest in Scotland (and elsewhere in the UK) in actively creating a social mix in regeneration areas was largely suspended after the global financial crisis of 2008, as governments grappled with simply increasing housing supply. Especially in places where parental choice is widely exercised, social mix at neighbourhood level is not a guarantee of more mixed school compositions (Lupton & Tunstall, 2008). Nevertheless, it would be good to see planning and regeneration strategies in Scotland much more actively consider their implications for the 'spatial school system', perhaps as part of the new 'local place plans', for which regulations and guidance are now being prepared following the Planning (Scotland) Act, 2019.
Second, schools could be better supported to look beyond their gates and work more directly with communities, especially where there are extant regeneration aims for disadvantaged areas. Cummings and Dyson (2007) provide some pointers based on empirical research, for example, while Riley (2013Riley ( , 2017 argues that school leaders should aim to understand the localities that shape young people's lives and become 'place makers'. This has common ground with Mowat's arguments about what it will really take to 'close the gap'. She argues for 'system leadership', including inter-school collaboration and shifting the emphasis from 'in-school' improvement towards 'between-' and 'beyond-school' improvement (Mowat, 2019b) and for strengthening networks between schools and their communities (Mowat, 2020). Indeed, learning programmes for school leaders promoted by Education Scotland (2020) emphasise working with communities to maximise positive educational outcomes. However, the available evidence on the use of the Attainment Scotland Fund does not suggest that there has been much development of outreach work so far (Scottish Government, 2019e).
Third, rather than only directing resources to individual schools, a place-based approach offers an alternative. Kerr et al. (2014) argue that 'new generation' area-based educational initiatives that are embedded in their localities and have a long-term perspective hold the promise of reducing risks and improving resilience as a route to redressing educational disadvantage. In Scotland, the initiative that most resembles this is Children's Neighbourhoods Scotland. Started in 2018, based in Glasgow's East End and inspired by the famous Harlem Children's Zone, it is aimed broadly at child poverty, with schooling one of several themes. In 2019 it received funding through the Scottish Government's Tackling Child Policy Strategy and is clearly an initiative worth watching as its approach is extended to other sites (Children's Neighbourhoods Scotland, 2019).
Fourth, optimism might be derived from the Scottish Government's emerging 'place principle' (Scottish Government, 2019c). The ambition is for a new cross-departmental focus on place, considered both as a resource for people and as an object of policy, reaching down to the 'community level'. However, buy-in so far seems to be mainly from those parts of government already concerned with communities (local government, housing and planning) and not, as yet, education.
Finally, achieving a place for place in education policy depends also on the academy. Although there is an international urban studies community with interests in education, recent contributions that have brought place and education together in the context of Scotland are scarce. The wider argument of this paper is that, in a country like Scotland that is riven by socio-spatial divisions, it is as important to understand place as it is to understand more prevalent education research themes such as poverty, parenting and pedagogy, and that all need to be conjoined in order to challenge the attainment gap. Notes 1. PISA (Programme for International Student Assessment) is an international survey of the performance of 15 year olds in mathematics, sciences and reading carried out every three years in OECD and other participating countries. 2. PISA scores, however, have been seized upon by the SNP's political opponents and were one of the reasons why education reform became a theme of the 2016 Scottish General Election.
In 2020, opposition parties forced the Scottish Government to widen its review of school education in the light of the publication of the latest PISA results and of falling exam pass rates (Seith, 2020). 3. The SIMD was refreshed in 2020 using the same methodology. 4. For example, they showed that difference in the average tariff score of school leavers from the most and least deprived 20% of areas was equivalent to four Higher level exam passes at grade A or three Advanced Higher passes at grade B. Tariff scores are a simple attainment measure that adds together the grades accumulated by students across all course levels and awards. 5. Education Scotland is an executive agency of the Scottish Government charged with improving the quality of Scottish education, including acting as the schools regulator. 6. This is not to say that head teachers should not be accountable nor that school staff should not be working to address inequalities. Elements of such approaches are supported in Scotland, including in teacher professional standards and training, with benefits accruing (e.g. Mowat, 2019a). 7. As the SNSA was being developed the Scottish Survey of Literacy and Numeracy (SSLN) was discontinued. Lindsay Paterson argued that the new data would be inconsistent and submitted that 'Scotland has no reliable method of monitoring the performance of schools in literacy and numeracy for the first time in almost 60 years' (as cited in Denholm, 2018), which he later repeated to the Education and Skills Committee of the Scottish Parliament (2019).