Characterisation of the geographical area

The geographical area of PIS is partially integrated in the sub-regions of Beira Baixa and Médio Tejo, both, in turn, being part of the Central Region of Portugal.⁷ However, between 1986 and 2015 it corresponded territorially to NUTS⁸ III of inland central Portugal and consisted of the municipalities of Mação, Vila de Rei, Proença-a-Nova, Sertã and Oleiros. In 2013, by way of a legal alteration, the PIS nomenclature became extinct.⁹ Nevertheless, as referred to by Helena ^{Cabeleira, Carolina Carvalho and Ana Madeira (2020)}, this designation “is still alive today in the language of the local population when they wish to refer to the territory of the community to which they belong” (p. 142) - a question of identity. Thus, the present analysis, whose chronological period is situated between the academic years of 2004/05 and 2014/05, maintains the PIS territorial unit, namely the five afore-mentioned municipalities, as its methodological reference. This option made it necessary to break down some of the statistical information on education, since the sources from the 2013/14 academic year adopt the new nomenclature and integrate the aforementioned municipalities in the sub-regions of Beira Baixa and Médio Tejo.¹⁰

It is worth noting that the geodesic centre of mainland Portugal is precisely in the municipality of Vila de Rei (Figure 1).

Source:^{Penteado (2012)}.

Figure 1 PIS map (Mainland Portugal) 

As already mentioned, the five constituent municipalities of PIS have a common identity based on the “genuineness of the rural” and the “importance of nature as a vehicle for development and territorial enhancement” (Roteiros mais Centro. Pinhal Interior Sul. Roteiro mais Centro, n. d.). It is an inland, essentially rural territory, subject to coastal economic growth and to population exodus; this, since the 1950s, has led to population ageing (calling demographic sustainability into question) and the desertification of the geographical area (^{Mateus, 2008}). On the other hand, as the population density is low, settlements are concentrated in small areas and parishes (^{Penteado, 2012}).

It is also significant that Gross Domestic Product (GDP) per capita in the five municipalities is below the national average (^{Martins, 2007}).¹¹ Furthermore, in 2011, the PIS recorded the lowest “activity rates” in the country (^{Portugal, 2012}). For all the above reasons - plus the low “economic, social and business attractiveness” of the area (^{Mateus, 2008}, p. 16), driving young people to migrate to urban centres in search of jobs and to pursue their studies, it is certainly not surprising that the resident population has low levels of schooling and qualification. This statement is further backed by the 2011 Census: of the 40705 inhabitants of PIS registered at the time, 5368 had no level of schooling and 16707 had only the 1st cycle of basic education (^{Portugal, 2012}).

This data emphasizes the coastal-inland dichotomy of the country; at the same time it reflects the decade-long inability of the political power to keep the resident population in the geographical area. The focus on the economic dimension in urban areas, leading to rural exodus, necessarily has repercussions for educational outcome indicators.

Analysis of educational outcome indicators

In this section some PIS educational outcome indicators are analysed; first, considering the universe of the geographical area and, subsequently, breaking down the data according to each of the five municipalities. It should be noted that, methodologically, the PIS geographical unit has been maintained, even though the sources referring to the 2013/14 academic year are in line with the new nomenclature and integrate the five municipalities in the sub-regions of Beira Baixa and Médio Tejo. The rationale behind the outcome indicators considered (enrolled pupils, retention and drop-out rates and number of educational establishments) also requires further explanation. On the one hand, since during the time period under study (2004/05 to 2014/15) many educational establishments had been suppressed, largely due to the decreased young population, it appeared logical to select the “enrolled pupils” and “educational establishments” indicators. On the other hand, the selection of the outcome indicator "retention and dropout rates" was not so obvious. The idea, in this case, was to try to assess whether the progressive decrease in the number of 1^st cycle of basic education schools over the decade, which in practice involved concentrating pupils in fewer but larger schools, supposedly, with better educational conditions (school centres), was in any way related to the retention and drop-out rates. From the outset it was acknowledged that this is a very complex issue, involving several variables. In this respect, José Maria ^{Azevedo (2014)} clearly explains:

It is not easy to isolate the effects of network and facility changes on the improvement of educational outcomes, let alone to establish cause and effect relationships. The most immediate changes in day-to-day school life - access to school meals, reduction in the number of schools operating on a dual system or use of specific facilities - are more perceptible than the impact on teaching, learning and outcomes. (p. 576).

Let us now begin by analysing the evolution of the number of pupils enrolled in the 1st cycle of basic education (Table 2).

The progressive reduction of this figure over the course of a decade is evident, and highly significant when considered as a whole. In other words, between 2004/05 and 2014/15 the number of pupils enrolled decreased by 40.1%. Of course, this data must be considered in light of other, namely demographic indicators. For example, the data shows that between 2001 and 2011, when the last two censuses with published data were conducted¹⁵, the resident population fell from 44803 to 40705 inhabitants (^{Portugal, 2012}), corresponding to a reduction of 9.1%.¹⁶ On the other hand, a trend towards an ageing population was observed in this universe. In fact, in the specified timeframe, the ageing rate¹⁷ was twice as high as the national figure (^{Branquinho, 2011})¹⁸ . As for age structure, it should also be noted that this decline was most prevalent among the young population.¹⁹ Furthermore, to better contextualise the reduction in the number of pupils enrolled, between 2001 and 2011, the fertility rate²⁰ in PIS was well below the national rate and even that of the Central Region (^{Branquinho, 2011}).

Over the decade from 2004/05 to 2014/15, in line with the decline in the number of pupils enrolled, a significant decrease was observed in the number of public educational establishments. Unfortunately, the source does not break down the school buildings into types. However, between 2004/05 and 2014/15, the percentage of school closures (30.1%) was largely due to school concentration measures and entailed the extinction of several 1st cycle of basic education schools with fewer than 21 pupils.

The retention and dropout rate indicator also calls for analysis. To what extent, for example, is it possible to establish a relationship between this indicator and the number of schools, considering that the extinction of unitarian schools and school concentration measures would supposedly bring better teaching and learning conditions? As mentioned in the reference to the research of ^{José Maria Azevedo (2014)}, this is a complex issue, involving several variables and requiring contextual data that is not available. It would be important, for example, to have an understanding of the social, economic, cultural, behavioural and motivational conditions of the pupils and their families and also to have information about the schools as an organisation. Returning to Table 2, it may be said that there is no cause and effect relationship between the progressive reduction in the number of schools over the decade and the observed variation in retention and drop-out rates. In fact, the fluctuation of these rates is quite clear, although the differentials are mostly negative, with the exception of the school years 2008/09, 2011/12, 2012/13 and 2013/14. These are, for the most part, the years when the rates equalled or exceeded the national average. Considering the retention and dropout rate, the National Education Council had already underlined a reversal of the decreasing trend recorded since 2001/02, for the whole country and from 2011 (Conselho Nacional de Educação. Estado da Educação 2013, ²⁰¹⁴). This situation was reversed as of the year 2014/15 (Conselho Nacional de Educação, Estado da Educação 2015, ²⁰¹⁶). It should also be noted that from 2010/11 the retention and dropout rates were aligned (PIS and mainland Portugal).

The three outcome indicators in Table 2 will now be analysed, but by breaking down the data per municipality.

During the decade under analysis, the number of enrolled pupils fell in all the PIS municipalities. However, the fall observed is not identical. For example, in Vila de Rei, the least populated municipality with the smallest surface area, the number of pupils decreased by 23.7%. This figure is well below the average of the five municipalities and, in fact, the lowest in the geographical area. On the other hand, in the most populous municipality (Sertã), a decrease of 28.4% was recorded, also well below the PIS average. In turn, the highest percentage was recorded in Proença-a-Nova (43.5%), the second most populous municipality. Although not exclusively, this data should be seen in light of the demographic evolution.²⁴

As regards the number of educational establishments, the pattern was characterized by a sharp reduction in all the PIS municipalities, with the exception of Vila de Rei. In fact, in the latter case, the network remained stable (only two public schools) throughout the decade. It should also be noted that at this time the private network expression in PIS was residual. For example, in Mação, there is no reference to any private school in the mentioned time span.²⁵ As anticipated in the analysis of Table 2, the important issue is to understand the extent to which the fall in the number of educational establishments was related to 1^st cycle of basic education schools (as an effect of the school concentration measures), given that the source used does not break down the building types. However, in the case of the Proença-a-Nova municipality, it is possible to overcome this difficulty by using another source, namely the Carta Educativa do Concelho de Proença-a-Nova, published in ²⁰¹⁵, which reports on the 1st cycle of basic education schools eliminated since the 1980s. Taking the established chronology into account, it may be noted that of the 13 schools closed down between the academic years of 2004/05 and 2014/15 (see Table 3) 12 were 1^st cycle of basic education schools.²⁶ In the last academic year mentioned, there were only two schools (school centres) of this type, one with nine classrooms and another with three classrooms.

On the other hand, the analysis of the retention and drop-out rates per municipality confirms a previously considered idea, when the PIS universe was taken into account, namely that there is no cause and effect relationship between the progressive decrease in the number of educational establishments over the decade and the observed variation in those rates. However, it is important to note a pattern, which mirrors that observed in Table 2, in the transition from the 2011/12 academic year to the following year, which is the significant increase in the retention and drop-out rates. After breaking down the data, it is possible to see that in all the municipalities this rate increased as the years under analysis passed; this phenomenon is particularly evident in Oleiros (1% to 6.4%) and in Vila de Rei (1.8% to 7.9%). There is no explanation for these figures. An interpretation of local data would be enlightening, for example, related to municipal school support policies.

In the following section, the options taken throughout the decade from 2004/05 to 2014/15 regarding the extinction of schools with fewer than 21 pupils and the subsequent school concentration will be analysed. In the debate, not only does the assumed relevance of the extinction of educational establishments emerge, it should be noted that in the municipality of Oleiros seven schools had fewer than ten pupils in the academic year of 2005/06, but also a number of solutions regarding school concentration - for example, related to their location (only in the municipality or also in some parishes) and size.

Measures/school concentration options

In the initial years of the 21st century, namely after the creation of the Special Programme for the Rehabilitation of the 1^st Cycle of Basic Education School Network in 2002, disapproval of the suppression of schools and the school concentration measures (especially in rural communities) became more expressive. The teachers' unions manifested these protests on behalf of the teachers, essentially arguing that school closures were based on economic rationality criteria. A more contextualised argumentation can be read in the pages of the Fenprof Magazine²⁷ in 2005 and 2006. In fact, the absence of a strategic vision for the 1st cycle of basic education was a main criticism in these protests. On the other hand, it should be noted that Fenprof did not advocate the maintenance of all schools with fewer than 21 pupils. It argued that while some of the small schools could be closed down, the progressive reduction of the school network should include the construction of medium-sized schools (Fenprof Magazine, November 2005)²⁸; otherwise, children in several parts of the country would have to spend long hours travelling on school transport. However, in February 2006, a new article brought to light the operational fragilities of many 1^st cycle of basic education schools, especially in rural areas. In addition to its identification of the difficulties experienced, the text proposed a rehabilitation of the small village school. In essence, it argued that the “maintenance of the village school can contribute to the healthy development of the child, avoiding constant travel and cultural uprooting” (Fenprof Magazine, February 2006, p. 20). Nevertheless, if it proved necessary to close some of the schools, this process, according to Fenprof, would have to include: i) the establishment of a consensus with the population; ii) the safeguarding of the children’s comfort and safety on school transport; iii) negotiations between the government and the trade union with regard to the rehabilitation of the school network; iv) the establishment of a dialogue between the trade union and the government on the rehabilitation of the school network; v) the establishment of a dialogue between the trade union and the government on the rehabilitation of the school network (Fenprof Magazine, February 2006). The following month, with the suggestive title “Government cuts schools. 1^st cycle rehabilitation can wait!”, the issue was back on the pages of the magazine. The aim was to contradict the idea, conveyed by the Ministry of Education, that the suppression of schools was related to reasons of an educational nature. This principle was questioned due to the fact that not only were schools being closed down, but also health centres, courts, maternity hospitals, among other public services - the real purpose was “to reduce the State to its minimum” (Fenprof Magazine, March 2006, p. 14). Thus, it was argued that the maintenance of small schools in some regions could help combat desertification. In this respect, João Serrão, an expert in Human Geography, makes some interesting remarks. In fact, on 29 August 2010, in an interview with the newspaper Público, suggestively entitled “Does the closure of a school kill a village or hasten the inevitable?”, he bluntly stated the following:

We can always say that the closure of a school only hastens the death of the location, because the lack of children would lead to the same result in a few years anyway. But this is only true because the Powers that be have long since completely abandoned the rural world. (Público, 29 August 2010).

He further alerted to the fact that the decision to close schools should be taken within the framework of multi-sectoral land planning policies, taking specific cases into consideration.²⁹ In this regard, in some situations, “the imbalance caused by the closure of a school can be fatal to years of work, carried out at a local level to combat desertification” (Público, 29 August 2010). He also warned against a pernicious idea, embraced by many municipalities: that desertification is combated by strengthening local government seats - the effect, he added, would be to suck in the rural population).

For the purpose of this research, it is important to understand how these issues were addressed by the municipalities, considering, in particular, those that integrated the area of PIS. In this respect, the municipal educational policy papers reveal important information. As already mentioned, they were privileged municipal policy instruments. Indeed, issues as relevant as the definition of the education and training supply network, the prospective analysis of the system, the management of educational facilities, the process of grouping schools, among others, are embodied in these policy papers.³⁰ Interestingly, they also permitted the participation of several local actors.³¹

The corpus of the source policy papers consists of four documents (relative to the following PIS municipalities: Oleiros, Vila de Rei, Mação and Proença-a-Nova).³² Most of these documents were approved between 2006 and 2007 (therefore, still in force); the Policy Paper of Proença-a-Nova is an exception as it was updated in February 2015.³³ As for the structure of these five instruments, they share a common framework³⁴, namely: i) concept, purpose and guiding principles of the policy paper; ii) socio-economic and demographic characterisation of the municipality; iii) characterisation and evolution of the educational system (including a prospective vision); iv) rehabilitation criteria for the school network; v) monitoring of the policy paper. Particular attention will be given to the prospective analysis and the rehabilitation criteria for the school network, in order to understand the rationale behind the adoption of the concentration measures.

The analysis began with the Carta Educativa do Concelho de Oleiros, approved in October ²⁰⁰⁶. One of the important topics is the prospective vision for the academic year 2010/11, namely supply-demand in the 1st cycle of basic education. In 2005/06, of the nine schools in the municipality, seven had fewer than ten pupils. The forecast for the 2010/11 academic year points to a reduction in the number of pupils and teaching establishments.³⁵ Consequently, given the expected demand, the Policy Paper proposes the following network rehabilitation: five educational establishments, with notable concentration in the municipality seat. In fact, the school centre that was already there would provide five classrooms, and it was estimated that the other schools, all located in different parishes and with a single class, would be occupied by between six and 15 pupils. It should be noted that in the academic year in which the Policy Paper was approved (2006/07), two schools were closed down (Mosteiro and Madeirã parishes). In the first case, the option was taken to transfer the school population to the municipal seat and, in the second, to the parish of Sobral. Yet even considering the rarefaction of the school population, belief in the school as the identity factor and “symbol of progress” of rural communities, to use the words of ^{Rui Canário (2008, p. 34)}, appears to be absent. The main focus was clearly the urban area of the municipality. However, it is worth noting the arguments used to support the school concentration measures. The Policy Paper states that it is important "to concentrate the supply of 1st cycle education in the most relevant population centres and with greater demographic dynamics"; and, at the same time, “to accompany this concentration with a rehabilitation of the supply, namely at the level of facilities, support infrastructures and school equipment” (Carta Educativa do Concelho de Oleiros, October ²⁰⁰⁶, p. 84). It is even understood, given the demography and the government's guidelines, that the closure of the1st cycle schools is inevitable. The fact that the growing urban concentration of the population is considered a strategic vehicle for the socio-economic development of the municipality is also relevant. Consequently, and to a large extent, reasons of this nature were behind the rehabilitation of the school network.

In the case of the municipality of Mação, the respective Policy Paper also emphasises the question of demography; this, unsurprisingly, is also the case in the other PIS municipalities. On the date of approval of the document (January 2007), there were eight 1st cycle of basic education schools³⁶; only one of the parishes did not have a school. However, it was generally understood that the situation in the parishes could change due to migratory movements within the municipality itself. There were some interesting assumptions regarding the rehabilitation of the school network. In fact, it was understood that the effective benefits of eliminating schools needed to be well assessed, since, given the territorial extent of Mação municipality (surface of 400.82 Km2), the generalised use of school transport would carry some degree of risk for the children. It also emphasised the following idea: “the number of pupils is an important indicator, but it should not be exclusive or determinant [for the closure of a school]” (Carta Educativa do Concelho de Mação, January ²⁰⁰⁷, p. 75). The proposal for the rehabilitation of the network contained in the Policy Paper echoes this purpose. Indeed, it proposes the extinction of two schools (with five and 12 pupils, respectively, in the parishes of Aboboreira and Vales de Cardigos). However, the intention is expressed to keep schools with 14, 16 and 18 pupils open.³⁷ The latter decision was made due to the foreseeable evolution of the school population (without significant changes in the medium term, according to predictions), as well as to the good quality of the facilities. On the other hand, it should be noted that the arguments for closing the school with five pupils were not only related to school attendance. Indeed, the fact that the children would have to travel every day to the School of Mação (municipality seat)³⁸ to attend curriculum enrichment activities was taken into consideration. Thus, with the extinction of the school in Aboboreira, the transfer of these pupils would only serve to anchor an existing practice. Moreover, the school of Mação would be assumed, “in the medium term, as a reference for the whole southern part of the municipality”, receiving children from various parishes (Carta Educativa do Concelho de Mação, January ²⁰⁰⁷, p. 79). The idea was not, therefore, to concentrate all the schools (and pupils) in the municipality seat, but into a more integrated solution to that observed in the Policy Paper of Oleiros, to constitute “three educational poles that would cover the whole territory homogeneously” (Carta Educativa do Concelho de Mação, January ²⁰⁰⁷, p. 79).

Moving on to an analysis of the Carta Educativa do Concelho de Vila de Rei, approved in December ²⁰⁰⁷, it should be noted that this is the least populated and smallest municipality (PIS geographical area). When the document was drafted, the only educational establishment in the centre of Vila de Rei admitted pupils from all the parishes (Vila de Rei, S. João do Peso and Fundada) and the various levels of schooling (from the 1^st cycle of Basic Education to secondary). This was the Escola Básica Integrada do Centro de Portugal [Integrated Basic School of the Centre of Portugal] where six classrooms were allocated to the 1st Cycle. In light of the above, it is not surprising that the school in rural areas is not referred to in the Policy Paper. On the other hand, the stability of the network in the decade from 2004/05 to 2014/15 is particularly notable (see Table 3). The estimated evolution of the school population for 2011 was based on the following assumptions: i) constant birth rate; ii) no migration. School attendance predicted to be between 157 and 176 pupils; higher than the 136 pupils registered (enrolled) in the 2006/07 academic year (see Table 3). Nor is it surprising that this figure was revised upwards, considering that the document in question conveys a degree of optimism; it even assumes “some capacity to attract the population” (Carta Educativa do Concelho de Vila de Rei, December ²⁰⁰⁷, p. 55). It is worth recalling that between 2001 and 2011, Vila de Rei was the only PIS municipality to register growth (albeit slight) in the resident population.³⁹ In this regard, the school network rehabilitation proposal considers the need for the Escola Básica Integrada do Centro de Portugal to have three more classrooms for the 1st Cycle. Furthermore, it is argued that the pre-school network should be integrated in that of the 1st cycle of basic education, ensuring the coordination and complementarity of the two levels. There is also a clear focus on curriculum enrichment activities and on the principle of “full-time school” (ensuring the pupils’ transport and meals). In short, the idea was to continue to provide public education to all the pupils of the municipality in a single building (the Escola Básica Integrada do Centro de Portugal).

The last document to be analysed is the Carta Educativa do Concelho de Proença-a-Nova, approved, as already stated, in ²⁰¹⁵. This document shows that the municipality adopted the guidelines of the Ministry of Education regarding the rehabilitation of the school network; however, it did not refrain from criticising the Ministry of Education for failing to deliver a contextualised educational policy. In fact, the municipality had followed the Resolution of the Council of Ministers no. 44/2010, of 1 June, which aimed to achieve three main objectives, namely: i) to adapt the school network to 12 years of schooling for all pupils; ii) to adapt the size and conditions of the schools to promote educational success and combat pupil dropout; iii) to promote the rationalization of school clusters in order to foster the development of a common educational project, coordinating the various education levels and cycles. However, it was also sensitive to other topics referred to in the aforementioned Resolution, namely, the eradication of isolated educational establishments, as well as the assumption that all the pupils used the spaces with a canteen, library and computer room, among others. In a more operational manner, the municipality of Proença-a-Nova took into account the demographic and development dynamics, the travel distances and the transport network, also considering use of the existing school facilities (Carta Educativa do Concelho de Proença-a-Nova, February ²⁰¹⁵). However, it criticised (without significant consequences) the most decisive criterion for closing schools, i.e., the number of enrolled pupils being lower than 21, as adduced in the following argumentation:

Beyond the rigidity of the number, which can sometimes be made more manageable, depending on supporting elements, the idea remains that we are attacking the effects and not the causes. The criterion appears to be inadequate both for the reality of most of the inland regions of the country and for the reality of the municipality, since, in many cases, only the school(s) in the parish seat can attain those numbers permanently. (Carta Educativa do Concelho de Proença-a-Nova, February ²⁰¹⁵, p. 119-120).

Two ideas are difficult to reconcile, as may be seen in the Policy Paper. On the one hand, the school network supply needs to ensure proximity, thus avoiding inconvenient travel; and, on the other hand, the concentration of the educational supply in urban areas (school centres) emerges as a means to ensure the quality of the educational services provided. The solutions adopted clearly overvalue the latter idea. In fact, it is argued that the reconfiguration of the school network should follow the demographic dynamics of the Proença-a-Nova municipality, which leaned towards increased urban concentration. Moreover, according to the prospective balance for 2021, the number of pupils enrolled in the 1st cycle would continue to decrease in the municipality, despite the increasing demand for education in urban centres. This trend is even regarded as a strategic vehicle for the development of the municipality. Consequently, the reconfiguration of the school network stems mainly from an economic rationale. In short, it would be a matter of: i) “concentrating the supply of 1^st cycle education in the most relevant population centre with the highest level of socioeconomic dynamics”; ii) rehabilitating the “supply, namely at the level of facilities, support infrastructures and school equipment” (Carta Educativa do Concelho de Proença-a-Nova, February ²⁰¹⁵, p. 126). As a result, the closure of one of the two existing school centres in the municipality is being considered, which would mean that only one (of the four existing parishes) parish would be left with a 1^st cycle of basic education establishment. The vision that identifies the existing school network - the symbolic effect of the school, as it were - as a factor in the creation of new dynamics in the rural environment is clearly absent. In fact, not only is the idea of progress and economic development associated with the urban space, but also the determinism of the minimum number of pupils to keep a school open is still predominant - between 2004/05 and 2014/15, 12 1^st cycle of basic education schools were closed in the municipality of Proença-a-Nova (for having fewer than 21 pupils). In short, what is recommended in the Policy Paper (and in that of Oleiros and Vila de Rei) clashes with the philosophy of the afore-mentioned National Programme for the Upgrading of the 1^st Cycle of Basic Education and Pre-School Education Network (2007). In other words, the idea was to combat overcrowding in populous urban areas and avoid rarefaction of the network in rural areas (^{Azevedo, 2014}).