Equitable school distribution

It is estimated that by 2030, more than 60 million children will not have access to primary school (UNICEF, n.d.). Although location and proximity to schools can affect accessibility and equity of education and children’s educational achievement, many schools are placed in a rather haphazard way (AlQuhtani, 2022). It means that schools are not built where needed, resources are not used where they can be most profitable, and some children may be excluded from education (Mendelsohn, 1996). Planning should therefore aim for an equitable distribution of schools, ensuring that the supply meets the demand for education, considering the resources available as well as future needs (Sylla and Tournier, 2013), so that as many children as possible get to school safely at minimum costs in terms of time and distance

Promising policy options

Investing in microplanning approaches and school mapping

School mapping is a set of techniques that have long been used to determine the location of schools in an efficient and equitable manner (Hite, 2011). It is a critical step in the planning process that allows national goals to be adapted to the diverse contexts at regional and district levels. The objective is twofold: optimizing the use of resources and providing equality of opportunity for all (Caillods et al., 1983). Beyond that, it is a tool for attaining education goals such as universal access and efficiency of the education system, to name a few (Hallak, 1977).

The preparation of a school map should result from back-and-forth exchanges between the national, regional, and local levels, but school mapping, like micro-planning, is essentially a regional and local exercise that reflects the specific needs and requirements of each area. The school mapping design process starts with formulating national objectives that are further “regionalized” and “sub-regionalized”. These sub-national objectives are those against which regional and local authorities elaborate proposals for school maps. The local level is responsible for drafting the most detailed school map. Thus, the aggregation of local school maps makes regional maps, which themselves add up to constitute the national map (Caillods et al., 1983). At the local level, school mapping essentially follows five steps:

1. Diagnosis (existing supply, current demand and disparities)
2. Establishment of norms and standards (e.g., type of construction) and criteria for catchment areas (e.g., population density)
3. Setting of school plan targets by area (e.g., the target participation rate for each area)
4. Projection of enrolment numbers
5. Proposals for school maps (Caillods et al., 1983; Hallak, 1977).

The preparation or revision of a school map should account for many factors (demographic, educational, geographical, etc.). For example, educational level networks need to be coordinated to rationalize the school map. Primary schools map must then be drawn first, followed by lower secondary and upper secondary school maps (Caillods et al., 1983).

Where several proposals emerge from applying the required criteria, it is imperative to weigh the cost-effectiveness of each option to determine the most desirable one (Caillods et al., 1983). Before the introduction of a school map, the following steps should be taken:

• Determine the administrative organization for school maps;
• Implement a pilot exercise in a representative district (especially for specialized education services);
• Train the staff (at all levels) responsible for drawing school maps and implementing them; and
• Set a timeline for all school map activities (Hallak, 1977).

More details on the recommended procedures for school mapping can be found here and here.

Cross-level participation

Although the central level is responsible for providing directives to lower administration levels, the bulk of the work should be done by regional and local levels. The latter’s participation is crucial in guaranteeing that the distribution of schools matches the needs of the populations and the specificities of each area (Caillods et al., 1983). In that, local involvement enriches the process with local experience and knowledge. This should be complemented by the constitution of interdisciplinary teams and parents’ and communities’ involvement (Bonner et al., 2011).

One-way, top-down communication where the central level would take full authority over school mapping or, conversely, situations where decisions are imposed by decentralized levels, entailing the risk of inequities across areas and mismatches with national goals, are to be avoided (Hallak, 1977). School mapping must be based on balanced relationships and an iterative process between the different levels of administration.

In practice, this is a difficult balance to strike. In some cases, lacking skills at the sub-national level and the complexity and cost of instruments used for school mapping can result in over-reliance on the central level expertise and resources to perform their tasks which, therefore, annihilates the benefits of micro-planning (Hite, 2011; Sylla and Tournier, 2013). There might also be instances where the central level is reluctant to devolve responsibilities (Sylla and Tournier, 2013). This might be the case because whichever level of government is responsible for schooling access often strongly influences decisions related to school location (Gershberg, 2014). These tensions can be mitigated with clear definitions, roles, goals, and intentions for participants at all levels (Hite, 2011). Besides, to avoid “piecemeal decisions”, responsibilities should be clearly defined to ensure greater accountability. Administrative reform can be considered to facilitate the process (Hallak, 1977: 135).

Backing up school mapping with instruments, tools, and data

School mapping effectiveness depends on the use of instruments and tools (Sylla and Tournier, 2013). Geographical Information Systems (GISs) are among the most powerful and cost-effective tools frequently used in this exercise: they “provide methods to capture and collect such [geographical] information, to store that information, to process it statistically and spatially, and finally to make that information available according to the needs of whoever requires it” (Mendelsohn, 1996: 4). Beyond school mapping, GIS can be used to, for example, investigate the relationships between space and educational outcomes or girls’ access to education as in Northern England and Uganda respectively (Mendelsohn, 1996).

For a GIS to work, educational planners would first need to define their expectations for the system in a way consistent with educational issues. This will help determine the types of products it is intended to deliver and increase ownership over them. Once the goals of the GIS have been set, the next step is developing the system, which can require specific expertise to help deal with technical issues (hardware, software, equipment) and implies building an adequate set of data (spatial and attribute information (e.g., demographics)) (Mendelsohn, 1996). Although spatial data gaps are still acute in developing countries, initiatives aiming to fill in this gap include UNICEF’s satellite imagery-based tool to map every school in the world and My School Today!, a citizen-monitoring effort that allows students and communities to geo-reference their schools (Sustainable Development Solutions Network, 2021; UNICEF, n.d.). Training and technical support in the development and use of GIS, as well as advocacy for its use, are key elements of its success (Mendelsohn, 1996).

In addition to GIS, geographically weighted regressions (GWR) are another tool that can be used to determine areas where policies may have the most impact. GWR is “an econometric regression method that allows for local variations in regression coefficients across space, where each local regression is weighted by the observations surrounding said location” (Gagnon and Vargas Mesa, 2021: 1). 

Differentiating urban and rural areas

Whether a school is in a rural or urban area poses different equity concerns across and within areas.

Given lower density in rural areas, private providers’ offer tends to be concentrated in urban areas, restricting parental school choice in the former (OECD, 2018). In OECD countries, the greater demographic pressure in urban areas compels supply expansion, whereas rural school networks are often characterized by excess capacity. If it is less likely an issue in cities, long distances to schools, which can affect children’s attendance and attentiveness, is a central concern when establishing rural school networks. These discrepancies call for differentiation when drawing school maps. As physical constraints (e.g., buildings and lands) are also particular, and the influence of municipalities is significant in urban contexts, a special organization for urban school map services should be established (Hallak, 1977). 

Inequities also exist within areas. In OECD countries, while rural areas are often heterogeneous in school locations, sizes, and educational offers, urban areas face higher risks of segregation as a result of a larger and more diverse educational offer and an uneven distribution of schools (OECD, 2018). Examples of initiatives undertaken to address segregation include the Copenhagen “Improving learning for all” program, which incentivizes immigrant students to attend native schools and vice versa. “Magnet schools” located in disadvantaged neighborhoods in the USA offer specialized courses and high-quality education to attract students from other areas. Both initiatives were supported by transport schemes to facilitate student mobility (OECD, 2018).

Transport arrangements

Distance-related norms and standards can be established to ensure more equitable locations. These can be (i) a minimum variability of access to schools, (ii) an average or maximum travel time to schools, or (iii) minimizing the maximum travel time to schools (Malczewski and Jackson, 2000). Increasing the number of schools can reduce the distance from many students’ homes but diminishes the chances for economies of scale. In sum, there may be trade-offs between efficiency and equity.

Transportation is an integral factor to school mapping which can help “reduce” distances, especially if the reorganization of the school network involves school closures, in which case arrangements must be made to avoid discontinued education (OECD, 2018). Some OECD countries have established thresholds in terms of distance from home to schools, along with factors like age and safety concerns, to determine students’ eligibility for school transport (OECD, 2018).

While some of these countries also have arrangements for students with special educational needs, long and unsafe roads to schools still pose a threat to education accessibility for millions of children in Asia and Africa (especially upper primary and secondary) (Agarwal, 2020). Curitiba, Brazil is an example of a city that invested in transport services to meet the needs of students with disabilities (Agarwal, 2020).

Transport is also a determinant of parental school choice. Improving connections can facilitate access to attractive schools, especially for disadvantaged students, and enhance diversity. While providing transport subsidies could be an option to achieve similar results, “policymakers need to carefully consider how they interact with school admissions criteria, in order to avoid incentivising parents to trade in school quality for lower transport costs” (OECD, 2018: 177).

Other policy options

Alternative modes of schooling

Boarding schools can be a solution to allow nomadic children to attend school or in situations where, for example, the population is too dispersed to open schools with sufficient enrollment (Caillods et al., 1983; Hallak, 1977). However, it is both costly and disruptive to children who are separated from their families (Behaghel, de Chaisemartin, and Gurgand, 2015). For this reason, “boarding facilities […] should be the exception than the rule” (Hallak, 1977: 15), at least for compulsory education, and adequate school mapping should allow avoiding this option.

Mobile schools constitute a flexible approach to providing educational opportunities for marginalized populations, notably in pastoral areas (e.g., Kenya). However, experience with mobile schools shows that it may not be embedded in a coherent policy framework, teachers often lack material and training support, and constant changes with children entering and leaving the school system can have negative consequences for their learning (UNESCO, 2010) (please consult School infrastructure for more information).

Distance learning can be a way to provide education to those who do not have access to physical facilities. For many children worldwide, this remains a costly option that may exacerbate existing inequalities and evidence about ICT-based education cost-effectiveness, particularly in rural areas, remains scarce. The methods used and the support to teachers and schools are of utmost importance in distance learning (OECD, 2018).

Adapting the organization of schools and school networks

Temporary solutions can be applied to resorb excess demand. These include redirecting students to schools with available places, using school physical space more efficiently, or extending facilities. The permanent or temporary nature of the latter will depend on the predicted duration of the increase in student enrollment (OECD, 2018). If an extension or new construction is not possible, and this is more likely the case in urban areas, an option would be multishift schooling, i.e., a system of rotating schedules among groups of students. However, its effects on the time and quality of instruction (e.g., reduced instruction time) must be carefully considered (see more on this topic in Double-shift schooling).

Several OECD countries have provided incentives or disincentives to schools to encourage the consolidation of their rural school networks. This involves closing schools that are deemed financially non-viable and relocating students to others. This approach seeks to achieve economies of scale through reduced cost per student. It can also help fight against segregation and inequalities, with positive consequences for students if they are redirected from low- to high-performing schools (OECD, 2018). Nevertheless, economies of scale may only be achieved up to a certain school size, beyond which returns to scale diminish. Besides, care must be taken to ensure higher transport costs do not outweigh savings. For parents and students, consolidation also means less inter-school competition and more limited school choice (OECD, 2018). Evaluation and consultation with relevant stakeholders are necessary to ensure the validity and appropriateness of consolidation.

School clustering and, more broadly, cooperation between schools, is one of the solutions to avoid or limit school closures. Cooperation between schools implies that the school network is not altered but that there is a sharing of resources and personnel, which would otherwise be too costly for a single school to provide. For example, small rural schools in France organize themselves in networks and cooperate in providing extracurricular activities, transport, and internet access (OECD, 2018).

In a school cluster, one school is designated as the lead and other nearby schools are considered satellites. In some cases, a single principal assumes leadership for all schools and the budget is joint; in others, leadership is located at the cluster’s central administration, but schools within the cluster operate autonomously (OECD, 2018). School clustering comes with challenges, including more complex organization of transport arrangements, longer travel times and scheduling coordination issues for staff assigned to multiple schools (OECD, 2018).