Is small beautiful? A scoping review of 21st-century research on small rural schools in Europe

Small rural schools in many countries have historically been viewed as less desirable than their larger urban counterparts, being treated less favourably in the policy arena and facing a risk of closure or amalgamation. Within Europe, they have been the focus of a range of research studies and have been defined in different ways, based mostly on the number of pupils enrolled (typically around 100), but also their geographical isolation or small number of staff. However, the last notable attempt to review the literature in this area was carried out over 10 years ago. Since then, there has been a large volume of research conducted in many European countries. This rigorous scoping review has brought together what we know from research (published in English since 2000) on small rural schools in Europe. The most common themes identified were school leadership, the importance of context (including education policy and school closures/amalgamations), and school-community relationships. Case studies and ethnographic qualitative methods were the most popular, with a stronger use made of theoretical frameworks since the previous review. Nonetheless, the current review found still significant gaps in the literature including an under-theorization of certain topics and a lack of research with children.


Introduction
According to Eurostat (2020), in total, just over a quarter (28%) of the population across 28 European countries live in what can be designated a rural area. These areas, however, vary markedly in socio-economic, geographical and educational terms. For example, while rural areas close to urban centres are likely to be economically dynamic, remote and sparsely populated areas present weaker economic growth and experience population decline and reduction in services, including education. It is precisely these areas which are served by small rural schools. Such small rural schools have been defined in different ways. While primarily related to the number of pupils enrolled, ranging from under 70 to under 140 for primary schools (Hargreaves, 2009), categorization has also included: the small numbers of staff employed in the school; the school's geographical isolation; and having principals with significant teaching commitments (Raggl, 2015). Regardless of definition, what is evident is that throughout Europe, small rural schools in remote and sparsely populated areas have closed or live under threat of closure/amalgamation due to a combination of factors, including the marketization of education and rationalization of services, a decrease in student numbers, difficulties in attracting and retaining staff (including principals), lack of funding, and diminishing resources Beach and Vigo Arrazola, 2020;Cannella, 2020;Kovács, 2012;Sigsworth and Solstad, 2005). In the areas in which schools have closed, children must travel long distances, which in turn has been shown to have an impact on their educational and emotional wellbeing (Beach et al., 2018), with such closures having considerable adverse impacts on the sustainability of the local communities (Kvalsund, 2009).
In contrast to this seemingly pan-European policy trajectory and its cultural, socio-economic educational impact, some authors (Gill, 2017;Hargreaves, 2009;Kvalsund, 2009;Raggl, 2020) have championed the cause of small rural schools, highlighting the potential they have for adding educational value, and the building of community engagement and social cohesion for sustainable change. For instance, because of their size and limited number of pupils, they have been recognized as giving teachers the opportunity to develop closer relationships with their pupils and attend to their individual needs (Aðalsteinsdóttir, 2008), as well as being able to provide unique and innovative educational opportunities (Raggl, 2020). In addition, they appear particularly well placed to contribute to and involve the local community in curriculum content and school practices (Autti and Baeck, 2021;Beach et al., 2019;Rothenburger, 2015), are locally accessible (Jones, 2009) and environmentally sustainable Gristy et al., 2020a).
In terms of understanding the scope of previous research on the state of small rural schools in Europe, relatively little work has been published (in English) which attempts to consolidate, frame and analyse the nature and findings of existing research publications. There are of course some noteworthy, if limited, examples where this has been attempted, most notably the work of Hargreaves et al. (2009) in the 2009 special edition of the International Journal of Educational Research. This publication included reviews of small rural school research in Sweden (Åberg-Bengtsson, 2009), Scotland (Dowling, 2009), England (Hargreaves, 2009), Finland (Kalaoja and Pietarinen, 2009) and Norway (Kvalsund, 2009). In the 10 years or so since publication of this special edition, an increased volume of rural educational research has been undertaken and published both in the countries originally covered, but also in those that were not. In writing this article, we believe there is a need to bring all this body of knowledge together, and to scope analytically what the research literature tells us on small rural schools in Europe. In order to undertake this task, we rigorously follow the methodology of a scoping review as outlined by Arksey and O'Malley (2005). Our analysis focuses on published research (in the English language) undertaken on small rural schools in Europe since the start of the new millennium.
Following an exposition of the methodology, the article highlights the most common areas that researchers have focused on, namely: the importance of the school context at the micro (e.g. the history of the school and its geography within a particular country/region) and macro level (e.g. the effects of national education and social policies and curriculum change); the relationship between schools and their communities; school leadership and management, including school collaborations and networks; the learning environment and multi-grade teaching; and equality of education and educational achievement. Our scoping of the literature also examines how the small rural school has been conceptualized in the last 20 years in empirical research. In the final section, challenges and opportunities for small rural schools in Europe that the literature has identified are summarized. Finally, we identify gaps in the literature and propose possible new research directions.

Methodology: scope of the review
Although there is not a universally accepted definition of a scoping review, its main trait involves giving a comprehensive overview of a broad topic (Peterson et al., 2017). To this end, we used the five steps outlined by Arksey and O'Malley (2005):

Identifying the research question
While in systematic reviews and meta-analysis the research questions tend to be very specific, a scoping review allows broader and more general questions (Peterson et al., 2017). Thus, we started the review with our main research question, which was: What does the existing literature tell us about small rural schools in Europe? A secondary research question was also raised: What theoretical frameworks and concepts have been used in research on small rural schools?

Identifying relevant studies
In order to be as comprehensive as possible, we adopted a search strategy involving different sources: electronic databases (i.e. British Education Index, ProQuest Education journals, Educational Administration Abstracts, International Bibliography of the Social Sciences (IBSS), Social Science Citation Index, and Google Scholar); reference lists; hand-searching of key journals (e.g. Journal of Rural Studies, European Educational Research Journal, International Journal of Educational Research); and social media used by academics (i.e. Twitter and ResearchGate). The search term was 'small rural school'. The various mechanisms for searching in our scoping study generated a total of 364 references (excluding duplicates), the majority of which were identified through the electronic databases.

Study selection
All the references were imported into EndNote, duplicates were removed, abstracts were read and screened for relevance, adhering to inclusion/exclusion criteria. Our selections were partial in so far as inclusion and exclusion criteria were decided based on our time and budget constraints, as well as our language limitations, and the purpose of the scoping review. Inclusion criteria comprised the following: • • peer-reviewed, scholarly journal articles, book and book chapters; • • written in English; • • published between 2000 and 2020; • • presenting either a review of research or empirical research data; and • • having a focus on small rural schools in a European country.
We did not prescribe a specific meaning of rurality or small rural schools, leaving the definition to the authors. We excluded discussion and practice articles (with no empirical data reported), theses, reports, conference proceedings and literature from non-European countries. Implementation or evaluation studies of educational interventions were also excluded. In addition, we excluded research studies conducted in one or more small rural school, but with no substantive interest in small rural schools as their primary focus; research could have been conducted in any type of school. Thus, many of the references excluded fitted in some of these categories (per order of volume): literature from non-European countries, literature that had not been peer-reviewed, discussion and practice articles, studies conducted in small rural schools but with no interest in small rural schools, and evaluations of educational interventions.
We acknowledge the pitfalls of excluding both literature written in other European languages and the grey literature. Regrettably, our knowledge of other languages was insufficient to appraise any academic literature, but we recognize that authors in southern and eastern European countries are less likely to publish in English than those in Scandinavian or English-speaking countries. Therefore, this is an unfortunate and clear limitation of this review. In terms of the grey literature, it has been argued that including grey literature avoids any publication bias. However, we believe it was important that all the literature included had been peer-reviewed.

Charting
After identifying the literature included in the review, we synthesized and interpreted the information in each of the references. In this phase, we created a Word table with each row corresponding to a particular study or a literature review article, and each column containing the following information:

Collating, summarizing and reporting the results
Having charted all the information collected, we have presented the findings in two ways. The first is shown in Table 1, where a numerical analysis of the literature is displayed, charting the number of references that touch on the different topics and theoretical concepts found. Many studies did touch on different issues, so they are counted in more than one row. Secondly, in the following sections, we present an account of the most common themes, exposing any differences or similarities between countries, describing a variety of findings, and identifying concepts used.

Small rural school: definitions
The definition of a small rural school varied within countries and studies. However, overall, small rural schools in the literature were mostly defined by the number of pupils enrolled. Small rural primary schools typically had fewer than 100 pupils. For post-primary schools, there was greater variation among countries but there were also fewer studies focusing on post-primaries. The 'small' number of pupils also had clear implications in terms of the size of staff and buildings/resources. Sometimes, the schools were characterized by their geographical location -for example, being remote or island schools -or by their situation -for example, being at risk of closure. Table 4 describes the different ways small rural schools have been defined or understood in some of the studies included here. As seen in these definitions, although the issue of size was often discussed and defined to some degree, the 'rural' dimension was addressed in only a few articles. No clear definition of 'the rural' or of 'what constitutes rurality' was attempted in most of the articles and book chapters reviewed. When the rural was explicitly defined, definitions were largely based on population density and inter-settlement distances (Gristy et al., 2020b). Thus, while in other fields of study (e.g. rural sociology, geographical studies, etc.) the notion of rurality has been extensively debated, in research about rural schools this issue has hardly been addressed, and there is instead an implicit assumption of what the 'rural' attribute is or signifies.

Sweden
A primary school situated in a small village with fewer than 100 pupils (aged 6-13) enrolled. A very small one has fewer than 50 pupils.
Autti and Hyry-Beihammer (2014) Finland A small school is a school with fewer than 50 pupils, usually located in a rural area, and typically 2 or 3 teachers teach different grades in the same classroom.
Bajerski (2020) Poland Definitions of small rural schools in Polish research and policy have included the low number of students (up to 30 or up to 70), their critical situation -i.e. being in danger of closing down, and their limited scopei.e. not providing the full education programme at the primary level.

Pettersson and
Ström (2019) Sweden A school located in a rural area with no more than 55 pupils enrolled and 1 to 8 teachers who teach in multi-grade classrooms.
Jones (2004) Wales Small rural secondary schools include schools in a rural catchment area for 11-18 year olds with a roll of fewer than 700 pupils, and schools for 11-16 year olds with a roll of fewer than 600 pupils.

Raggl (2019) Austria
Small rural primary schools are schools located in a rural area with fewer than 50 pupils which have mixed-grade classes.

Emergent themes
In Table 1, we summarize the key themes and concepts identified in the scoping review, before going on to address each of these in more detail. Leadership was the most common topic (with 33 references), especially in empirical studies (rather than literature reviews), and it included articles focusing on the role of the principal, school boards, school collaborations or networks, curriculum development and change, and teachers' professional development. The methods employed in these studies were also the most diverse, with about half employing qualitative methods and the other half using a mixed method design. The following most common topics were context (31 references), school-community relationship (28) and learning environment (23). The topic of context embraced a range of themes, including school closures/amalgamations or the threat of closure, national education policies and rurality. Most of the studies included in this topic employed qualitative designs, and most of the literature reviews included in this review (7 out of 11) referred to this topic. School-community relationship was also a very popular topic, which included the role of school in the community, the impact of school closures on the community, and community engagement with schools. The learning environment topic mostly examined the practice of multi-grade teaching, but also included the theme referring to information and communications technology (ICT) and distance education, and teacher collaboration. The least explored topic was equality of education and academic achievement (12 references), which was referred to in nearly half of the literature reviews (including a research synthesis) (5 out of 11), but only a handful of studies (7 out of 36) dealt with these issues.

The importance of context
The importance of context and the diversity of small rural schools have been emphasized in many of the studies included in this review (see Table 1) Hargreaves et al., 2009). In these studies, schools are recognized as distinctive at the micro-level (i.e. associated to a specific local spatial geography and history) but also connected to large political, economic and social structures (Beach and Vigo Arrazola, 2020).
Regarding the macro-level, much of the literature has described similar changes in Europe's rural areas, with small rural schools closing or amalgamating. Some countries, particularly many Scandinavian countries, used to have a high proportion of small rural schools by the mid-20th century, but this number has significantly diminished since then. For instance, in northern Finland, between 1990 and 2010, three out of four small rural schools have closed (Autti and Baeck, 2021). A similar case is evident in non-Scandinavian countries such as Poland, where the number of rural primary schools dropped by nearly 40% between 1990 and 2012, resulting in the closure of nearly 6000 schools (Bajerski, 2020).
Indeed, school closures and schools under imminent threat of closure have been the focus of studies across Europe: in Finland (Autti and Hyry-Beihammer, 2014;Karlberg-Granlund, 2011), Norway (Aasland and Søholt, 2020), Latvia (Koroļeva et al., 2017;Supule, 2019), Czech Republic (Kučerová and Trnková, 2020), Sweden (Cedering and Wihlborg, 2020;Villa and Knutas, 2020) and England (Walker, 2010). The reasons behind school closures in most countries are cited typically as financial (i.e. small rural schools are too costly to run in comparison to larger schools). In several countries, authors Bajerski, 2020;Solstad and Karlberg-Granlund, 2020) have reported that there is a policy bias which presumes that rural schools are somehow deficient, that multi-age teaching in small rural schools is 'inferior' to age homogenous teaching in larger schools, and that the way to improve rural education is making schools bigger. Significantly, these prejudices seemingly exist despite little or no hard evidence that pupils in small schools do worse than others in larger ones (Solstad and Karlberg-Granlund, 2020).
Regardless of these global trends, national education policies and socio-economic situations are different in the various European countries where studies are situated. In post-Soviet and postsocialist countries, after the fall of the communist regimes, there was a determination to decentralize policymaking, but this occurred in different ways and with different consequences (Gristy et al., 2020a). Kovács (2012) situated their case study of a small village school in the context of the Hungarian countryside, which was experiencing a process of counter-urbanization and also the process of impoverishment and ghettoization in that particular rural area (i.e. ethnic segregation of Roma and non-Roma people). So, in the studied school, demographic decline led to a sharp drop in enrolment, which was fought by stimulating the attraction of non-Roma children from a neighbouring village (where its own school comprised 70% of Roma children). The phenomenon of attracting pupils from outside the local area, however, has been explored in other case studies. In some jurisdictions like England, wealthy and middle-class city dwellers relocate to rural areas (Hillyard and Bagley, 2015). One further reported case was an English school which was under threat of closure and located in a remote geographic location (Walker, 2010). The author concluded that 'small schools opting for a niche market as a school-management policy jeopardize the future of the school if this turns away local parents on a scale that cannot be compensated for by likeminded parent choosers coming in from further afield' (Walker, 2010: 724).
Some studies have compared small schools located in different countries and they have found differences but also similarities. For instance, differences in terms of stakeholders' attitudes regarding school closures were found between Latvia and Norway. Thus, whereas in Latvia the majority of school administrators held a positive opinion about small schools, with schools and municipalities agreeing that school closures and mergers should be avoided, this opinion was not shared in Norway (Koroļeva et al., 2017). Beach and Vigo Arrazola's (2020) cross-national meta-ethnography comparing rural schools in Sweden and Spain focused on the similarities between these two countries in terms of the marketization of the education system, characterized by policies of private choice, decentralization and competitive funding. The authors identified two types of schools in Spain's and Sweden's rural areas: 'communitas schools' and 'magnet schools'. Whereas communitas schools were considered as having developed strong links to families and the local community, with teachers, parents, pupils and community members working hard together for their survival, magnet schools had not and were getting half or more of their pupils from middle-class families from outside the schools' local area. This finding mirrors Kovács' (2012) and Walker's (2010) studies with magnet schools exploiting their existing forms of capital and the market conditions they operated in.
Even within the same country, studies have distinguished between different ruralities and types of rural areas. Raggl (2020: 213) emphasized the plurality of small rural schools 'and how much they differ according to their location, their numbers of pupils and teachers, and whether or not they have to fight for their existence'. In fact, in her study, she distinguished between different types of small rural schools in Austria: very small and remote single-teacher schools (with fewer than 20 pupils); small 2-teacher schools located approximately 30 minutes' driving time away from a small town (with 20-30 pupils); and small Montessori schools located near urban centres (with 20-35 pupils). She found that the different types of schools faced different challenges, with teachers facing more isolation and fluctuation in the very small remote schools (Raggl, 2019). Similarly, Beach et al. (2019), drawing on a materialist spatial geographic theoretical perspective (Massey, 1994), in which space is understood as fluid and formed through socio-spatial and material practices, emphasized the importance of context and the plurality of rural schools.
Moreover, recent local social histories impact on the schools' current realities, as highlighted in Hillyard and Bagley's (2019) case studies of two English rural schools. When focusing on one of these case studies, Hillyard and Bagley (2013) revealed how the interpersonal issues (high staff turnover and the legacy of a former principal), political-economic (the village shifting local 'squirearchy') and spatial dynamics (post-war expansion and the situation of amenities) together influenced the fact that the school lacked a full-time principal for several years.

The school-community relationship
One of the advantages traditionally associated with small rural schools has been their strong connection with the local community. Indeed, teachers and principals interviewed in various studies (Karlberg-Granlund, 2019;Trnková, 2009;Tsiakkiros and Pashiardis, 2002) identified as a strength the positive relationships between their schools and the community. In addition, there was evidence from different countries that small rural schools had extended roles within their communities. For instance, Supule (2019) found that 78% of rural schools they surveyed in Latvia were an open space to community members (e.g. school premises were used for sports and hobby activities).
It was found that studies that have focused on this topic often employed sociological concepts such as social cohesion and social capital, particularly Putnam and Feldstein's (2003) distinction between bonding, bridging and linking social capital. Bonding social capital refers to close connections within particular social groups, bridging social capital refers to more distant connections between different groups, and linking social capital refers to connections between different levels of power hierarchies. Such conceptual distinctions were used in four of the studies included here (Autti and Hyry-Beihammer, 2014;Hemming, 2018;Karlberg-Granlund, 2019;Kovács, 2012). Authors found that small rural schools boosted and created bonding and bridging social capital within the local community, but this could potentially operate in a negative/exclusionary way (Bagley and Hillyard, 2014). For instance, both small rural faith schools in Hemming's (2018) English study were able to meaningfully engage with their local communities (bonding and bridging social capital), particularly through strong links with local parish churches. At the same time, commuters and newcomers to the local area who came in order to take advantage of the school were seen as disrupting the village character, and social divisions emerged. In addition, community cohesion can sometimes be experienced by teachers and principals as a mechanism of restrictive control in a small rural school, due to the tensions caused by social proximity (Karlberg-Granlund, 2019).
Some schools have been labelled as 'community active schools', particularly in the Nordic countries (Kalaoja and Pietarinen, 2009), but also in other countries like Ireland (Gill, 2017) or Austria (Raggl, 2019). This label originates from Solstad's (1997) distinction between 'community-ignorant' schools (i.e. acting like isolated islands in the community), 'community-passive' schools (i.e. using the local community for teaching and learning) and 'community-active' schools (i.e. integrating teaching and learning in the local context and playing an active role in the community). In Norway, community-active schools are most frequently found among small schools (Karlberg-Granlund, 2019). In Austria, another study also found that some of the schools that took part were community-active schools, but the majority of principals/teachers that took part did not engage in village activities outside their work and did not live in the village where they worked either (Raggl, 2019). In Finland, a review concluded that half of the small rural schools in the country were considered community-active schools (Kalaoja and Pietarinen, 2009).

The learning environment
Within the learning environment theme, authors mostly focused on multi-grade teaching and the use of ICT. When talking about multi-grade teaching or classes, we are referring to classrooms where children of different ages and grades are taught together. As highlighted in Gristy et al. (2020a), this has been named in different ways by different authors in different countries, including multi-age, mixed age and incomplete teaching.
Multi-grade teaching has been considered as both a necessity and an opportunity, and teachers, principals and educators have shown opposite views regarding its value (Raggl, 2015). Deunk and Maslowski (2020), for example, found that of 26 principals interviewed who mentioned this issue, 9 thought multi-grade teaching was too difficult and should be avoided if possible, 6 felt multigrade classrooms were managed well by their highly qualified staff, and 11 were not sure and thought multi-grade classrooms were hard to manage but they could work well. Difficulties highlighted by teachers in different studies are to do with evaluation (i.e. difficulty and doubts in finding the right method) (Domingo Peñafiel and Boix Tomàs, 2015) and planning, which becomes more complex in multi-grade classrooms and involves a higher workload, putting pressure on teachers with less experience (Raggl, 2015). In addition, it requires sufficient training preparation, which many teachers do not have, and most school books and materials are oriented according to age-homogenous classes (Raggl, 2015).
In our sample, there were two studies and a research synthesis that focused on the teaching strategies used in multi-grade classrooms. In Spain, Domingo Peñafiel and Boix Tomàs (2015) found that the educational methodology used by teachers in multi-grade classrooms in three rural schools was a mixture of promoting participation among all of the students and the development of active educational strategies (where the focus was the type of activities around which the students' work and the teachers' lesson plan evolves). Regarding the spatial organization, this depended on the activities being developed, while the time organization differed between the three case studies observed (from a rigid schedule to an autonomous development of activities). The textbook was the main educational material in two of the schools; in the other, the textbook was just one tool among others. Hyry-Beihammer and Hascher (2015) also found that teaching practices in multigrade classrooms varied widely in Austria and Finland, and were associated with teacher personality, subjects and teaching situations. However, some common practices were identified. The subjects of science, religion and art were usually taught as whole-class teaching, whereas mathematics and (native) language were commonly taught using the parallel curriculum, which means that both grades are taught the same but have different assignments. Peer tutoring was also found to be common, and the authors identified two forms of peer tutoring: spontaneous peer tutoring (i.e. when children help others unprompted) and guided peer tutoring (i.e. as a reaction to the teacher's guidance). One of the most challenging tasks identified by teachers in both countries was how to meet the needs of different learners. Teachers believed individual differentiation to be essential because of the wide range of differences between learners. The authors identified two main strategies: (a) one reducing or overcoming the heterogeneity of students as much as possible (using parallel curricula, curriculum rotation and whole-class teaching); and (b) one capitalizing on the heterogeneity of the students, but also reducing teaching demands (using peer tutoring, personal work plans or free work). In their research synthesis of studies from Austria, Finland, Spain and Switzerland, Smit et al. (2015) concluded that mixed-age learning is most effective when: • • teachers support peer-and cross-age tutoring, but also work with groups and give instruction at the same time; • • tutoring supplements teacher instruction but does not replace it; • • the age-gap in peer tutoring should not be more than four years; • • teachers know where the student is in the learning process in order to guide instruction and provide feedback; and • • teachers implement cross-age, open-ended tasks to enable simultaneous learning and assessment.
The use of ICT in the learning environment of small rural schools was the focus of a few studies. In Italy, Cannella (2020) identified two models of lessons using ICT in three small mountain and island school networks: (a) the Common Learning Environment scenario, where two or more classrooms work together on a common project-based activity using different technological settings; and (b) the shared lesson, where two multi-age classrooms work together using a videoconferencing system. However, for ICT to be used positively in the classroom, she found that four principles needed to be in place: appropriate technology infrastructure, good school organization, well-trained teachers and appropriate curriculum content and organization. She concluded: [S]chools can work with local authorities and with other small rural schools in order to create a national network to include those on mountains and small islands. If a community of schools with similar features can work together, they will be able to offer a rich learning environment to the students living in those isolated parts of the country. (Cannella, 2020: 232) Similarly, in Finland, Hilli (2020aHilli ( , 2020b focused on virtual learning environments (VLE) in three small rural schools, and found that the VLE enabled teacher collaboration, but it needed the same digital infrastructure (and advanced digital competence to use it) in all the schools, which proved difficult. For instance, most teachers and students lacked the digital competence to interact through the learning management system. Teachers appreciated the collaboration, as it supported their professional development and reduced their professional isolation.

School leadership
Some studies (six included in this review) have focused on the figure of the school principal in the small rural school. Given the heterogeneity of small rural schools within and between countries, the situation of their principals is also diverse. Deunk and Maslowski (2020) interviewed principals in the north of the Netherlands, and some were teaching heads while others were multi-school principals (i.e. part-time principals of two or three schools). In their study, teaching heads had opposite reactions to this role. Some felt it was stressful, while others considered teaching as an enrichment. For multi-school principals, the advantage was cooperation between teams of different schools, whereas the disadvantage most mentioned was a lack of time for educational leadership. In Cyprus, Hadjitheodoulou Loizidou and Fokaidou (2020) tried to understand the complexity of the leadership experience of a novice active principal in a small rural school in her first year of headship. They used Clarke and Hollingsworth's (2002) Interconnected Model of Teacher Professional Growth as a framework, which emphasizes teachers' active role in their own professional development and learning. According to this model, teachers' knowledge is a result of active and meaningful change, which is achieved through four different domains: the personal domain (i.e. knowledge, belief and attitudes); the external domain (i.e. external sources of information or stimuli); the domain of practice (i.e. professional experimentation); and the domain of consequences (i.e. prominent outcomes related to practice). In their case study, Hadjitheodoulou Loizidou and Fokaidou (2020) found that the new principal felt isolated and hopeless in the face of challenges in the external domain, having never worked in a small rural school, but 'commitment and care in interaction overcame challenges through reflection on definitions of situation' (174).
The type of culture intrinsic in school leadership was explored by Wildy et al. (2014). In two case studies in Iceland and Australia, they concluded that both principals were contextually literate (having an appreciation of place) but both schools were characterized by a culture of acceptance (where the conditions they find themselves in are accepted but not questioned). That contrasts with a culture of inquiry, which entails wilfully engaging in critical discourse to produce personal meaning and mutual understanding. They argued that to promote such culture, principals in small rural schools need continuing professional support and adequate preparation. There have been, indeed, calls for principals and teachers to be provided with appropriate training and professional development focused on leading and teaching in rural schools. Principals and teachers often appear to come unprepared for the context of working in small rural schools.
Leadership in some countries did not only refer to principals, but also to other figures, like the joint superintendent as in Norway (Forfang, 2020), or school boards (Deunk and Maslowski, 2020), and collaboration between different roles within the education system was found to be crucial for the school functioning and even its survival.
Another topic within the area of leadership that was the focus of certain studies in this review is curriculum change and development. Studies demonstrated the importance of cooperation and collaboration between principals and others within the school community to carry out school reform and curriculum change successfully. This task was identified as difficult when the numbers of staff are very low (one-, two-or three-teacher schools), leading to obvious practical difficulties in implementing any change (Wilson and McPake, 2000). Thus, school size was found to influence local curriculum development processes, as development processes cause a lot of administrative work for a few members of staff (Autti and Baeck, 2021). There were also differences between and within countries. While in Norway 'principals transferred the responsibility of implementation and the efforts to work with the new curriculum to the teachers' (Germeten, 2011: 21), principals in Scotland appeared to see themselves as 'part of the teaching team' and led from within the team rather than from outside to implement four new initiatives (Wilson and McPake, 2000). Germeten's (2011) study explored the role of school principals and others in implementing curriculum change in northern Norway, and argued that in small rural schools, curriculum reform may not succeed unless the school owners, on the municipality level, involve themselves in the actual implementation process. Managing school reforms and curriculum planning presuppose a professional and official system, not single principals working in schools as though the enterprise was theirs alone. (21-22) Wilson and McPake (2000) found that the type of management activities used by principals to implement four initiatives in Scotland constituted a distinctive small schools' management style (i.e. situational school management). Their implementation strategy could be defined as: quick audit; realistic planning for achievable goals; implement through networking and support from internal/external sources; and move on to next initiative.
In Finland, curriculum planning and change has involved considerable challenges for small rural schools, particularly in terms of 'the management of change, the organization of teaching, professional development of the teachers and the co-operation between the school and the surrounding community' (Kimonen and Nevalainen, 2001: 172). In their case study of a small rural school, Kimonen and Nevalainen (2005) identified a series of school characteristics that enabled the principal and the school community to be successful in implementing rapid curriculum changes, including informal relationships between staff and pupils; radical changes in the teachers' models of thinking and action; and pupils', parents' and community members' positive attitude regarding the pedagogical changes in the school.
In different countries, some small rural schools have coped with financial and educational challenges (caused by their size) by forming networks and collaborations, clusters or federations, which work in different ways. In Norway, collaborations between schools appeared endemic in some regions' educational infrastructure, as in Forfang's (2020) study, the schools in a network of three rural municipalities had a joint superintendent, and the schools were governed through networking and inter-organizational collaboration. In Scotland, small rural schools have formed clusters or loose collaborative groups, which appear to have been largely successful in terms of teamwork, sharing expertise with staff in other schools, minimizing duplication of effort, providing new opportunities for staff development, and reducing professional isolation (Dowling, 2009). Research in England has also highlighted the benefits of clustering for small rural schools, as national policy has encouraged small schools to work together and form networks and federations (Hargreaves, 2009). However, although collaborations between small rural schools are mostly perceived as beneficial in expanding opportunities for teaching and learning, principals can also conceive them as extra work. Jones (2009), for instance, found that the principals they interviewed in England discussed collaboration 'almost exclusively in terms of the time, money and workload involved' (129).

Equality of education and academic achievement
Equality of education and academic achievement was less explored, with seven empirical studies included in this review focusing on these issues. Some of these studies have focused or touched on special education in small rural schools. In general, studies have indicated the advantages of small rural schools in supporting children with special education needs. For instance, participating principals in a case study of five small rural secondary schools in Wales argued that one of their strengths was high levels of support for pupils with special educational needs (Jones, 2004). Similarly, teachers in Cyprus who participated in a mixed methods study felt that pupils' special needs can be recognized more quickly in small schools (Tsiakkiros and Pashiardis, 2002). In Sweden, all the teachers interviewed in three case studies (Pettersson and Ström, 2019) considered the rural school as 'a good learning environment for all students', and they had access to a special educator working in the municipal administrative centre, whom they could contact when they needed a consultation. In this study, teachers asked for support rather than advice when they contacted special educators, and they described the consulting discussions as a professional dialogue/exchange. In contrast with all these studies, findings from a survey of teachers in Iceland (Aðalsteinsdóttir, 2008) did not indicate that small schools were particularly well equipped regarding giving attention to pupils with special needs. According to the author, this appeared to be due to a variety of reasons, including staff being poorly trained and isolated; poor resources in terms of curriculum facilities; or being unprepared to exploit the potential advantages of small schools.
Educational achievement and performance in small rural schools has been examined in several literature reviews and a few empirical studies. Literature reviews and studies present a mixed and complex picture in relation to comparing educational performance between small rural schools and large urban schools. This unclear picture is highlighted in Dowling's (2009) scoping review regarding research on rural Scottish education, and Åberg-Bengtsson's (2009) review of Swedish research found that there was no evidence that performance is better in larger schools (but it seemed to be rather the other way round). While teachers in Tsiakkiros and Pashiardis' (2002) study in Cyprus found the performance standards of pupils in small schools lower than those of pupils in large schools, in Poland, according to Bajerski's (2020) review, studies conducted after the 1998 reforms have shown that deep differences in the educational achievements of pupils from rural and urban schools have been eliminated. However, inequalities have emerged between peripheral and central areas (major urban agglomerations) 'that not only differ because of the availability of social, cultural and financial capital resources, but also of actual accessible opportunities to select schools and extracurricular activities' (Bajerski, 2020: 133).

Conceptual and theoretical frameworks
The importance of theory in this area of work has been emphasized by authors such as Gristy (2020), who argued that applying theory to research on education in rural areas, particularly in case studies, can enhance the beneficial aspects of the footprints of research. In this review, more authors appeared to mention specific theoretical concepts and theories when contextualizing and analysing the findings of their research than what was found over 10 years ago, in Kvalsund and Hargreaves' (2009) review.
Nevertheless, many research studies reviewed here did not appear to start from a theoretical standpoint, but they often used theory a posteriori (rather than a priori) to make sense of the data collected. For example, some appeared to use some form of grounded theory approach, which was not necessarily clearly articulated, with findings subsequently thematically analysed and descriptively reported. In Table 5, we show the range of theories mentioned and applied to the research studies; most of which we have referenced previously in relation to our discussion around prevalent themes. Theories around the concept of social capital were the most used, followed by Gidden's structuration theory and Lefebvre's conceptualization of space, and stemmed from different disciplines, including sociology, political theory, geography and education. While some authors clearly used theory to 'inform research questions, the nature of sample, the selection of methods and the interpretation of the data' (Kvalsund and Hargreaves, 2014: 44), a few others merely mentioned the theory as a 'genuflection', rather than meaningfully using it in their research. Among many others specified in Table 5, examples of authors using theory to inform their research questions and interpret their data were Beach et al. (2019), drawing on materialist spatial geographic theory to explore how rural schools related to the local place, and Hillyard

School closures
• School-community relationship

Sweden
System leadership and (geographical, organizational, cultural, institutional, social and cognitive) proximity (Knoben and Oerlemans, 2006) Forfang (2020) School collaboration/ networking Norway Traditional curriculum theory (Lundgren, 1979) Germeten ( (2020), who applied Bourdieu's field theory to a case study of a small rural school in England, particularly focusing on the school's changing relationship with the local community. The theoretical background of the different studies often depended on the topic the author(s) focused on, and the methodology used. In contrast with studies conducted previously, most studies included in this review used a bottom-up approach and a life-world perspective (Hargreaves et al., 2020). Thus, they are studies where the school and the local community were examined at the local level where 'the everyday interplay between actors, conditions and processes of place -nature, production, culture -and education takes place' (Hargreaves et al., 2009: 141). In addition, a few (particularly those studying the learning environment) focused on the micro level.

Concluding discussion: omissions, challenges and opportunities
In reviewing the literature on small and rural schools in England over 10 years ago, Hargreaves (2009) argued that research on this topic had been pragmatic rather than theory-driven and conducted in response to government policies, from the perspective of the school rather than its local community. In the last two decades, while many studies still do not specifically frame their research within a theoretical paradigm, several researchers around different European countries have used theoretical frameworks to make sense of the data collected, including social capital and structuration theory (shown in Table 5). Nevertheless, many of the topics and themes specified in this review remain relatively under-theorized, and they could arguably benefit from a wider use of theoretical perspectives. Similarly, despite a few notable exceptions Gristy, 2020;Hillyard and Bagley, 2013;Rothenburger, 2015), there remains a lack of conceptualization of the rural; of what constitutes rurality in these studies.
Despite the undoubted progress in research, our review found that some significant gaps identified in previous reviews have not yet been filled. For instance, Hargreaves et al. (2020) have recently referenced the under-representation of children's voices in this area. Although there were a few studies that did focus on pupils' and young people's views (or included them to some degree) (Kvalsund, 2004;Öhrn and Beach, 2019;Raggl, 2015), most of the research included in this review focused on adults' perspectives (i.e. mostly teachers', principals' or parents' views). Just like a decade ago (Kvalsund, 2009), it would appear little effort has been made to meaningfully engage children in research and explore children's agency, despite the centrality of children within the past, present and future situation of small rural schools (Kvalsund, 2019). This is in contrast to educational research in general, in which research embracing student's voice has proliferated since the 1990s and early 2000s (Cook-Sather, 2018). Thus, we believe that the field would significantly benefit from European researchers meaningfully involving and engaging children in research, actively seeking their views/perspectives.
Noticeable is the way in which the relationship between schools and their communities has become an increasingly popular theme in most of the countries of the studies included in this review. Most of the research across Europe has utilized qualitative (often ethnographic) case studies to reflect the situation of small rural schools. Thus, as Åberg-Bengtsson (2009) advocated over 10 years ago and Hargreaves et al. (2020) observed more recently, many of these studies have taken 'a broad perspective from the inside and below'; that is, taking a closer look at the situation of small rural schools and the people/communities involved in them.
Thus, based on the findings from this scoping review, research on small rural schools in the last two decades appears to be dominated by qualitative methodologies, and a dearth of quantitative studies (going beyond the use of surveys) is apparent. Although ethnographic and qualitative methods continue to be key for the development of the field, there is also a case for large-scale quantitative studies to be carried out, either at the national level or regionally, as well as comparative studies between rural/urban and small/large schools. Such studies could potentially substantiate some of the claims made about the positive aspects of small rural schools (Hargreaves et al., 2020). In addition, more sophisticated methods such as meta-analysis, inter-generational designs, and geographical studies of closures and their effects on everyday population movement (Gristy et al., 2020a) could potentially strengthen the field. However, having identified these gaps, this scoping review is limited by the previously mentioned language factor. Thus, these conclusions need to be taken with the caution that we do not know what others in some countries have published in different European languages other than English.
From the scoping review of research, we have seen how small rural schools have been under threat of closure and amalgamation in many European countries. Indeed, in the last 40 and 50 years, many have closed. Research has indicated that they are facing many challenges across all these countries and regions, including: • • financial pressures; • • staff's intense (and often unmanageable) workload (including teaching principals having a double job); • • high staff turnover and difficulties in recruiting and retaining qualified staff; • • dwindling pupil enrolment; • • professional isolation of teachers and principals; • • inadequate infrastructure; • • challenges in delivering a wide-ranging curriculum; and • • pedagogical challenges of multi-grade teaching.
Despite the identified challenges, practical and effective solutions to these challenges have been proposed. These include: • • training and professional development programmes for teachers and principals, which are tailored for working in the small rural school environment; • • suitable material resources; • • school collaborations, clusters and networks to reduce professional isolation and boost schools' resources; and • • distance learning, virtual learning environment, and the use of ICT (when the right infrastructures are provided, i.e. appropriate digital structures and teacher and pupils' digital competence to use them).
In several studies, despite all the challenges mentioned, teachers and principals have expressed being satisfied with working in a small rural school (Karlberg-Granlund, 2019; Raggl, 2015), and some believed that it was a good learning environment for all pupils, including those with special needs (Pettersson and Ström, 2019). In fact, small rural schools have also been shown to provide multiple opportunities to teachers, principals, pupils and the communities they serve. Firstly, small rural schools are possibly well placed to provide a 'place-based curriculum', which emphasizes children's relationship with nature and promotes local history and culture . Secondly, small rural schools appear better able to form strong links to their communities, as well as utilize the communities' potential to enrich the educational experience of their pupils, which in turn builds social capital and social cohesion (Karlberg-Granlund, 2019;Raggl, 2019;Wildy et al., 2014). Thirdly, small rural schools have shown their potential to identify and address pupils' individual needs (Tsiakkiros and Pashiardis, 2002). Finally, small rural schools have been used as a testing ground for innovation (Raggl, 2020). In relation to all of the above, and returning to the title of our paper, small would in some ways still appear to be beautiful! Certainly, what our scoping review has also highlighted is the enormous diversity of small rural schools in Europe both within and between countries. While some of these schools appear to have found a niche and are coping well with the challenges of being small and rural, others appear to be struggling with declining numbers and shrinking resources. Judging by the literature reviewed, the sustainability and value of small rural schools depend not simply on their specific local circumstances and national policy context, but also on the lens of the individuals whose views are gathered -principals, teachers, policymakers, pupils, families and wider communities. Thus, although small rural schools in different European countries and rural areas appear to have many commonalities, their situation remains complex, diverse and socio-politically contested.